Re: [asterisk-users] Speech recognition in asterisk using google voice API
Hey Zaf, Just checking the Google Speech Recognition package again and I can't see WolframAlpha.agi file. I check all of your projects on Git hub but can't find wolframalpha.agi. Please let us know what the URL is. Thanks, Bruce On Thu, Jan 12, 2012 at 2:49 PM, Lefteris Zafiris zaf@gmail.com wrote: On 01/12/2012 05:50 PM, Danny Nicholas wrote: Two more offerings - #1 - add DTMF parameter so function can be stopped by pressing a digit or digits other than * or # - #2 - add an option to silence the beep. If you were using this in an IVR and wanted to say press 1 or say help for help, silencing the beep before recording would (IMO) make the rendering sound more professional/less mechanical. Both features added: - Usage - agi(speech-recog.agi,[lang],[timeout],[intkey],[NOBEEP]) Records from the current channel untill the timeout (set to 10 seconds by default, -1 for no timeout) is reached or the interrupt key (# by default) is pressed. If NOBEEP is set, no beep sound is played back to the user to indicate the start of the recording. There is now also the option to enable SSL for encrypted communication between your pbx and the google voice server. Updated code can be found here: https://github.com/zaf/asterisk-speech-recog/tarball/master Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Two more offerings - #1 - add DTMF parameter so function can be stopped by pressing a digit or digits other than * or # - #2 - add an option to silence the beep. If you were using this in an IVR and wanted to say press 1 or say help for help, silencing the beep before recording would (IMO) make the rendering sound more professional/less mechanical. -Original Message- From: asterisk-users-boun...@lists.digium.com [mailto:asterisk-users-boun...@lists.digium.com] On Behalf Of Lefteris Zafiris Sent: Saturday, January 07, 2012 6:22 AM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [asterisk-users] Speech recognition in asterisk using google voice API On 01/07/2012 09:34 AM, Bruce B wrote: Added two new features to the script: Timeout value and speechdata type. *exten = s,n,agi(speech-recog.agi,en-US,3000,phoneNumb)* - Will listen for 3 seconds and sanitize return as a single number without any spaces in between. This helps when one reads phone number in format 415-554-2323 and google returns, 415 554 2323 as result which is not very usable. *exten = s,n,agi(speech-recog.agi,en-US,2,string)* - Will listen for 20 second and return result as provided by Google untouched. It would be great to see them in future versions as I seem to need them dearly in a real life scenario. Updated script attached. -Bruce Thank you Bruce for the testing and the suggestions. Both features added in the script. Timeout can now be set by the user, also -1 means no timeout and the recording keeps going till # is pressed. Space gets stripped between digits, this is now the default behavior and there's no need to determine the 'speechdata' type. The updated code can be found here: https://github.com/zaf/asterisk-speech-recog/tarball/master Next on my TODO list is to make use of the asterisk speech recognition API (https://wiki.asterisk.org/wiki/display/AST/Speech+Recognition+API) This will make the application actually usable for real case scenarios and not a proof of concept as it is now. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
On 01/12/2012 05:50 PM, Danny Nicholas wrote: Two more offerings - #1 - add DTMF parameter so function can be stopped by pressing a digit or digits other than * or # - #2 - add an option to silence the beep. If you were using this in an IVR and wanted to say press 1 or say help for help, silencing the beep before recording would (IMO) make the rendering sound more professional/less mechanical. Both features added: - Usage - agi(speech-recog.agi,[lang],[timeout],[intkey],[NOBEEP]) Records from the current channel untill the timeout (set to 10 seconds by default, -1 for no timeout) is reached or the interrupt key (# by default) is pressed. If NOBEEP is set, no beep sound is played back to the user to indicate the start of the recording. There is now also the option to enable SSL for encrypted communication between your pbx and the google voice server. Updated code can be found here: https://github.com/zaf/asterisk-speech-recog/tarball/master Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
On 01/07/2012 09:34 AM, Bruce B wrote: Added two new features to the script: Timeout value and speechdata type. *exten = s,n,agi(speech-recog.agi,en-US,3000,phoneNumb)* - Will listen for 3 seconds and sanitize return as a single number without any spaces in between. This helps when one reads phone number in format 415-554-2323 and google returns, 415 554 2323 as result which is not very usable. *exten = s,n,agi(speech-recog.agi,en-US,2,string)* - Will listen for 20 second and return result as provided by Google untouched. It would be great to see them in future versions as I seem to need them dearly in a real life scenario. Updated script attached. -Bruce Thank you Bruce for the testing and the suggestions. Both features added in the script. Timeout can now be set by the user, also -1 means no timeout and the recording keeps going till # is pressed. Space gets stripped between digits, this is now the default behavior and there's no need to determine the 'speechdata' type. The updated code can be found here: https://github.com/zaf/asterisk-speech-recog/tarball/master Next on my TODO list is to make use of the asterisk speech recognition API (https://wiki.asterisk.org/wiki/display/AST/Speech+Recognition+API) This will make the application actually usable for real case scenarios and not a proof of concept as it is now. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Does sox have more features on a Debian system than RHEL? Is that why it won't work on RHEL? Cheers, On Wed, Jan 4, 2012 at 6:42 PM, Lefteris Zafiris zaf@gmail.com wrote: Fresh code is out! The use of sox can be now optionally enabled by the user if the system has a recent version of the program (won't work in RHEL/Centos 5) This is done by editing the script and setting the variable 'use_sox'. When sox is used the audio gets normalized, low frequency noise (100Hz) is removed and also possible DC offset is corrected. Those are supposed to improve the recognition results(?). The settings are still a bit experimental, feel free to play with them and report what settings improved your results. get the new version here: https://github.com/downloads/zaf/asterisk-speech-recog/asterisk-speech-recog-0.3.tar.gz Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
On Fri, 6 Jan 2012 20:46:14 -0500 Bruce B bruceb...@gmail.com wrote: Does sox have more features on a Debian system than RHEL? Is that why it won't work on RHEL? RHEL's 5 version of sox is really old and outdated. The command syntax and the switches are totally different compared to recent versions of sox. Anyway I'm not sure audio normalization and the rest we use sox for is really needed. My tests so far didn't show any improvements in detection rates. Keep in mind that all this is still WIP and the option to use sox is more for testing than for serious use. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Thanks. I have been testing Aastra phones with SIP and had great results. I am testing my cell phone now and sometimes get -1 for id, status, utterance, and confidence. What does that mean? Cheers On Fri, Jan 6, 2012 at 9:40 PM, Lefteris Zafiris zaf@gmail.com wrote: On Fri, 6 Jan 2012 20:46:14 -0500 Bruce B bruceb...@gmail.com wrote: Does sox have more features on a Debian system than RHEL? Is that why it won't work on RHEL? RHEL's 5 version of sox is really old and outdated. The command syntax and the switches are totally different compared to recent versions of sox. Anyway I'm not sure audio normalization and the rest we use sox for is really needed. My tests so far didn't show any improvements in detection rates. Keep in mind that all this is still WIP and the option to use sox is more for testing than for serious use. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
NVM. I explored the code and see the logic. I had sox = 1 so it was failing on RHEL. To report, my cell phone from a PRI gets same confidence level just like SIP. Building my control app now. Should make my life much easier while driving. Thanks again :-) -Bruce On Fri, Jan 6, 2012 at 10:50 PM, Bruce B bruceb...@gmail.com wrote: Thanks. I have been testing Aastra phones with SIP and had great results. I am testing my cell phone now and sometimes get -1 for id, status, utterance, and confidence. What does that mean? Cheers On Fri, Jan 6, 2012 at 9:40 PM, Lefteris Zafiris zaf@gmail.comwrote: On Fri, 6 Jan 2012 20:46:14 -0500 Bruce B bruceb...@gmail.com wrote: Does sox have more features on a Debian system than RHEL? Is that why it won't work on RHEL? RHEL's 5 version of sox is really old and outdated. The command syntax and the switches are totally different compared to recent versions of sox. Anyway I'm not sure audio normalization and the rest we use sox for is really needed. My tests so far didn't show any improvements in detection rates. Keep in mind that all this is still WIP and the option to use sox is more for testing than for serious use. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Added two new features to the script: Timeout value and speechdata type. *exten = s,n,agi(speech-recog.agi,en-US,3000,phoneNumb)* - Will listen for 3 seconds and sanitize return as a single number without any spaces in between. This helps when one reads phone number in format 415-554-2323 and google returns, 415 554 2323 as result which is not very usable. *exten = s,n,agi(speech-recog.agi,en-US,2,string)* - Will listen for 20 second and return result as provided by Google untouched. It would be great to see them in future versions as I seem to need them dearly in a real life scenario. Updated script attached. -Bruce On Fri, Jan 6, 2012 at 11:03 PM, Bruce B bruceb...@gmail.com wrote: NVM. I explored the code and see the logic. I had sox = 1 so it was failing on RHEL. To report, my cell phone from a PRI gets same confidence level just like SIP. Building my control app now. Should make my life much easier while driving. Thanks again :-) -Bruce On Fri, Jan 6, 2012 at 10:50 PM, Bruce B bruceb...@gmail.com wrote: Thanks. I have been testing Aastra phones with SIP and had great results. I am testing my cell phone now and sometimes get -1 for id, status, utterance, and confidence. What does that mean? Cheers On Fri, Jan 6, 2012 at 9:40 PM, Lefteris Zafiris zaf@gmail.comwrote: On Fri, 6 Jan 2012 20:46:14 -0500 Bruce B bruceb...@gmail.com wrote: Does sox have more features on a Debian system than RHEL? Is that why it won't work on RHEL? RHEL's 5 version of sox is really old and outdated. The command syntax and the switches are totally different compared to recent versions of sox. Anyway I'm not sure audio normalization and the rest we use sox for is really needed. My tests so far didn't show any improvements in detection rates. Keep in mind that all this is still WIP and the option to use sox is more for testing than for serious use. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users speech-recog.agi Description: Binary data -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
On 01/04/2012 07:51 AM, Bruce B wrote: And with recent version 14.3.2 I get: /usr/local/bin/sox FAIL formats: no handler for file extension `flac' -- speech-recog.agi: /usr/local/bin/sox failed: 512 -- SIP/-002eAGI Script speech-recog.agi completed, returning 0 Regards, On Wed, Jan 4, 2012 at 12:43 AM, Bruce B bruceb...@gmail.com wrote: Very interesting. I just tried to get it to work but it complains about sox. Probably you used a different version of sox? *PBX-*CLI /usr/bin/sox: invalid option -- -* */usr/bin/sox: invalid option -- n* */usr/bin/sox: invalid option -- o* */usr/bin/sox: -r must be given a positive integer* * -- speech-recog.agi: /usr/bin/sox failed: 512* I am using: *Package sox-12.18.1-1.el5_5.1.i386 * Thanks, Note to self: Never release anything asterisk related without testing on RHEL/Centos 5 Thank you for reporting this. I have replaced sox with flac and it seems to work now on older platforms too (tested on Centos 5 with asterisk 1.4). You can get the updated code here: https://github.com/zaf/asterisk-speech-recog/tarball/master Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
this looks great - is there any chance of coverting the googletts.agi to use flac as well ? Julian On 4 January 2012 09:06, Lefteris Zafiris zaf@gmail.com wrote: On 01/04/2012 07:51 AM, Bruce B wrote: And with recent version 14.3.2 I get: /usr/local/bin/sox FAIL formats: no handler for file extension `flac' -- speech-recog.agi: /usr/local/bin/sox failed: 512 -- SIP/-002eAGI Script speech-recog.agi completed, returning 0 Regards, On Wed, Jan 4, 2012 at 12:43 AM, Bruce B bruceb...@gmail.com wrote: Very interesting. I just tried to get it to work but it complains about sox. Probably you used a different version of sox? *PBX-*CLI /usr/bin/sox: invalid option -- -* */usr/bin/sox: invalid option -- n* */usr/bin/sox: invalid option -- o* */usr/bin/sox: -r must be given a positive integer* * -- speech-recog.agi: /usr/bin/sox failed: 512* I am using: *Package sox-12.18.1-1.el5_5.1.i386 * Thanks, Note to self: Never release anything asterisk related without testing on RHEL/Centos 5 Thank you for reporting this. I have replaced sox with flac and it seems to work now on older platforms too (tested on Centos 5 with asterisk 1.4). You can get the updated code here: https://github.com/zaf/asterisk-speech-recog/tarball/master Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- Julian Lyndon-Smith IT Director, Dot R Limited I don’t care if it works on your machine! We are not shipping your machine!” The kangaroo dances: http://www.youtube.com/watch?v=MAWl5iYOaUg -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
On 01/04/2012 04:07 PM, Julian Lyndon-Smith wrote: this looks great - is there any chance of coverting the googletts.agi to use flac as well ? Julian In googletts.agi we get the voice data from google in mp3 and we convert it in a format that asterisk can read and playback (slin). If we store it in flac asterisk wont be able to read it natively and we would have to convert it each time we want to play it back to the user. In the speech recognition script we have to convert the voice data in flac before sending it to google because that's the accepted format. Is there some particular reason you want the googletts.agi data in flac? Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
the only reason is that I didn't want to have to install sox. Lazy. that's all ;) Just another piece of software to find and install running on amazon ec2, is the best thing to download the source and compile sox ? Thanks Julian On 4 January 2012 14:18, Lefteris Zafiris zaf@gmail.com wrote: On 01/04/2012 04:07 PM, Julian Lyndon-Smith wrote: this looks great - is there any chance of coverting the googletts.agi to use flac as well ? Julian In googletts.agi we get the voice data from google in mp3 and we convert it in a format that asterisk can read and playback (slin). If we store it in flac asterisk wont be able to read it natively and we would have to convert it each time we want to play it back to the user. In the speech recognition script we have to convert the voice data in flac before sending it to google because that's the accepted format. Is there some particular reason you want the googletts.agi data in flac? Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- Julian Lyndon-Smith IT Director, Dot R Limited I don’t care if it works on your machine! We are not shipping your machine!” The kangaroo dances: http://www.youtube.com/watch?v=MAWl5iYOaUg -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
On 01/04/2012 04:24 PM, Julian Lyndon-Smith wrote: the only reason is that I didn't want to have to install sox. Lazy. that's all ;) Just another piece of software to find and install running on amazon ec2, is the best thing to download the source and compile sox ? Thanks It should be on your distro repos already. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
nope :( On 4 January 2012 14:29, Lefteris Zafiris zaf@gmail.com wrote: On 01/04/2012 04:24 PM, Julian Lyndon-Smith wrote: the only reason is that I didn't want to have to install sox. Lazy. that's all ;) Just another piece of software to find and install running on amazon ec2, is the best thing to download the source and compile sox ? Thanks It should be on your distro repos already. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- Julian Lyndon-Smith IT Director, Dot R Limited I don’t care if it works on your machine! We are not shipping your machine!” The kangaroo dances: http://www.youtube.com/watch?v=MAWl5iYOaUg -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Note to self: Never release anything asterisk related without testing on RHEL/Centos 5 Thank you for reporting this. I have replaced sox with flac and it seems to work now on older platforms too (tested on Centos 5 with asterisk 1.4). You can get the updated code here: https://github.com/zaf/asterisk-speech-recog/tarball/master Lefteris Zafiris Works beautifully. Amazing job Lefteris. Thanks. The best result I got in probability was 0.9725632 by saying, hello. I think there is some non-phonetic logic built-in as well. I tried, 1, 2 and I got 0.86534226 in accuracy. While I tried 1, 2, 3, 4, 5 I got, 0.97256315. Probably Google sees the pattern?! What are some of the other tricks (if any) or consideration that one should make while creating a strong speech recognition enabled IVR? Best, -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Does anyone know what languages are supported? -Original Message- From: Bruce B bruceb...@gmail.com Sender: asterisk-users-boun...@lists.digium.com Date: Wed, 4 Jan 2012 13:25:18 To: Asterisk Users Mailing List - Non-Commercial Discussionasterisk-users@lists.digium.com Reply-To: Asterisk Users Mailing List - Non-Commercial Discussion asterisk-users@lists.digium.com Subject: Re: [asterisk-users] Speech recognition in asterisk using google voice API -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Wow - nice! A few quick questions: 1. How long can the recording be for translation? 2. Any limitation on how much text the return (transcribed) variable can hold? 3. Any commercial / terms of use limitations? From: asterisk-users-boun...@lists.digium.com [asterisk-users-boun...@lists.digium.com] On Behalf Of Bruce B [bruceb...@gmail.com] Sent: Wednesday, January 04, 2012 1:25 PM To: Asterisk Users List Subject: Re: [asterisk-users] Speech recognition in asterisk using google voice API Note to self: Never release anything asterisk related without testing on RHEL/Centos 5 Thank you for reporting this. I have replaced sox with flac and it seems to work now on older platforms too (tested on Centos 5 with asterisk 1.4). You can get the updated code here: https://github.com/zaf/asterisk-speech-recog/tarball/master Lefteris Zafiris Works beautifully. Amazing job Lefteris. Thanks. The best result I got in probability was 0.9725632 by saying, hello. I think there is some non-phonetic logic built-in as well. I tried, 1, 2 and I got 0.86534226 in accuracy. While I tried 1, 2, 3, 4, 5 I got, 0.97256315. Probably Google sees the pattern?! What are some of the other tricks (if any) or consideration that one should make while creating a strong speech recognition enabled IVR? Best, -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
On Wed, Jan 4, 2012 at 8:47 PM, Michelle Dupuis mdup...@ocg.ca wrote: Wow - nice! A few quick questions: 1. How long can the recording be for translation? At the moment the recording timeout is set at 15sec. I haven't tested yet the max length of voice data ta google accepts (all this voice recognition stuff is undocumented). I have read that it is between 10-20 seconds but havent really went to test this yet. On my todo list is to add the option to cut the sound data in smaller chunks before sending them to google and get rid of the recording length limitations. 2. Any limitation on how much text the return (transcribed) variable can hold? This better be answered by the astsrisk devs but empirically talking i have loaded in dialplan variables really big chunks of text (like the complete gpl license) without having any problems. 3. Any commercial / terms of use limitations? This is a gray area at the moment. Voice recognition is undocumented in google's API and i guess not officially supported yet. I hope it gets covered by the general TOS of google services: http://www.google.com/accounts/TOS Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
On Wed, Jan 4, 2012 at 8:27 PM, isr...@gmail.com wrote: Does anyone know what languages are supported? For sure english and spanish, since its undocumented i don't have a complete list yet. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Works beautifully. Amazing job Lefteris. Thanks. The best result I got in probability was 0.9725632 by saying, hello. I think there is some non-phonetic logic built-in as well. I tried, 1, 2 and I got 0.86534226 in accuracy. While I tried 1, 2, 3, 4, 5 I got, 0.97256315. Probably Google sees the pattern?! What are some of the other tricks (if any) or consideration that one should make while creating a strong speech recognition enabled IVR? Google accepts sound files at any sampling rate (up to 44.1kHz) so if you can use some wideband codec ( eg g722) It can greatly improve the sound quality and the detection rates. For now the script supports 8kHz and 16kHz sampling rates for recording and it can be set by editing the scripts user defined parameters ( the variable $samplerate). Anything that improves the recording sound clarity will help, a good phone, low background noise level etc. I have also read that normalizing the recording and setting the gain to -5 db improves detection rates. I m experimenting with this at the moment and there will be some new code soon (as soon as i get sox working in RHEL/Centos 5 :P ). Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
On 1/4/2012 2:26 PM, Lefteris Zafiris wrote: Works beautifully. Amazing job Lefteris. Thanks. The best result I got in probability was 0.9725632 by saying, hello. I think there is some non-phonetic logic built-in as well. I tried, 1, 2 and I got 0.86534226 in accuracy. While I tried 1, 2, 3, 4, 5 I got, 0.97256315. Probably Google sees the pattern?! What are some of the other tricks (if any) or consideration that one should make while creating a strong speech recognition enabled IVR? Google accepts sound files at any sampling rate (up to 44.1kHz) so if you can use some wideband codec ( eg g722) It can greatly improve the sound quality and the detection rates. For now the script supports 8kHz and 16kHz sampling rates for recording and it can be set by editing the scripts user defined parameters ( the variable $samplerate). Anything that improves the recording sound clarity will help, a good phone, low background noise level etc. I have also read that normalizing the recording and setting the gain to -5 db improves detection rates. I m experimenting with this at the moment and there will be some new code soon (as soon as i get sox working in RHEL/Centos 5 :P ). This is really spectacular. Thanks. I'm running Fedora 15, so I can use flac or sox. Any reason to prefer one over the other? sean -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
wow i just tried in hebrew and i'll say just 1 word WOW On Wed, Jan 4, 2012 at 9:48 PM, sean darcy seandar...@gmail.com wrote: On 1/4/2012 2:26 PM, Lefteris Zafiris wrote: Works beautifully. Amazing job Lefteris. Thanks. The best result I got in probability was 0.9725632 by saying, hello. I think there is some non-phonetic logic built-in as well. I tried, 1, 2 and I got 0.86534226 in accuracy. While I tried 1, 2, 3, 4, 5 I got, 0.97256315. Probably Google sees the pattern?! What are some of the other tricks (if any) or consideration that one should make while creating a strong speech recognition enabled IVR? Google accepts sound files at any sampling rate (up to 44.1kHz) so if you can use some wideband codec ( eg g722) It can greatly improve the sound quality and the detection rates. For now the script supports 8kHz and 16kHz sampling rates for recording and it can be set by editing the scripts user defined parameters ( the variable $samplerate). Anything that improves the recording sound clarity will help, a good phone, low background noise level etc. I have also read that normalizing the recording and setting the gain to -5 db improves detection rates. I m experimenting with this at the moment and there will be some new code soon (as soon as i get sox working in RHEL/Centos 5 :P ). This is really spectacular. Thanks. I'm running Fedora 15, so I can use flac or sox. Any reason to prefer one over the other? sean -- __**__**_ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/**mailman/listinfo/asterisk-**usershttp://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
On Wed, 04 Jan 2012 14:48:22 -0500 sean darcy seandar...@gmail.com wrote: This is really spectacular. Thanks. I'm running Fedora 15, so I can use flac or sox. Any reason to prefer one over the other? sean We have to convert the voice data to flac format before sending them to google, this can be done by both sox and flac encoder. For now the script uses flac encoder for compatibility with older distros (mainly RHEL 5). Sox is a bit more flexible and also gives you the option to edit the sound data (normalizing, changing levels etc). Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Fresh code is out! The use of sox can be now optionally enabled by the user if the system has a recent version of the program (won't work in RHEL/Centos 5) This is done by editing the script and setting the variable 'use_sox'. When sox is used the audio gets normalized, low frequency noise (100Hz) is removed and also possible DC offset is corrected. Those are supposed to improve the recognition results(?). The settings are still a bit experimental, feel free to play with them and report what settings improved your results. get the new version here: https://github.com/downloads/zaf/asterisk-speech-recog/asterisk-speech-recog-0.3.tar.gz Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[asterisk-users] Speech recognition in asterisk using google voice API
Hello, I have written an agi script that uses google voice API for voice recognition. The script records from the current channel untill the pound key (#) is pressed or the timeout (15 seconds) is reached. The recording is send over to google speech recognition service and the returned text string is assigned to a channel variable. More info and dialplan examples can be found in the README file: https://raw.github.com/zaf/asterisk-speech-recog/master/README The script is available here: https://github.com/zaf/asterisk-speech-recog The code is still young and not roughly tested so comments, suggestions and bug reports are more than welcome. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Very interesting. I just tried to get it to work but it complains about sox. Probably you used a different version of sox? *PBX-*CLI /usr/bin/sox: invalid option -- -* */usr/bin/sox: invalid option -- n* */usr/bin/sox: invalid option -- o* */usr/bin/sox: -r must be given a positive integer* * -- speech-recog.agi: /usr/bin/sox failed: 512* I am using: *Package sox-12.18.1-1.el5_5.1.i386 * Thanks, On Tue, Jan 3, 2012 at 9:42 PM, Lefteris Zafiris zaf@gmail.com wrote: Hello, I have written an agi script that uses google voice API for voice recognition. The script records from the current channel untill the pound key (#) is pressed or the timeout (15 seconds) is reached. The recording is send over to google speech recognition service and the returned text string is assigned to a channel variable. More info and dialplan examples can be found in the README file: https://raw.github.com/zaf/asterisk-speech-recog/master/README The script is available here: https://github.com/zaf/asterisk-speech-recog The code is still young and not roughly tested so comments, suggestions and bug reports are more than welcome. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
And with recent version 14.3.2 I get: /usr/local/bin/sox FAIL formats: no handler for file extension `flac' -- speech-recog.agi: /usr/local/bin/sox failed: 512 -- SIP/-002eAGI Script speech-recog.agi completed, returning 0 Regards, On Wed, Jan 4, 2012 at 12:43 AM, Bruce B bruceb...@gmail.com wrote: Very interesting. I just tried to get it to work but it complains about sox. Probably you used a different version of sox? *PBX-*CLI /usr/bin/sox: invalid option -- -* */usr/bin/sox: invalid option -- n* */usr/bin/sox: invalid option -- o* */usr/bin/sox: -r must be given a positive integer* * -- speech-recog.agi: /usr/bin/sox failed: 512* I am using: *Package sox-12.18.1-1.el5_5.1.i386 * Thanks, On Tue, Jan 3, 2012 at 9:42 PM, Lefteris Zafiris zaf@gmail.comwrote: Hello, I have written an agi script that uses google voice API for voice recognition. The script records from the current channel untill the pound key (#) is pressed or the timeout (15 seconds) is reached. The recording is send over to google speech recognition service and the returned text string is assigned to a channel variable. More info and dialplan examples can be found in the README file: https://raw.github.com/zaf/asterisk-speech-recog/master/README The script is available here: https://github.com/zaf/asterisk-speech-recog The code is still young and not roughly tested so comments, suggestions and bug reports are more than welcome. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech recognition in asterisk using google voice API
Hi there, I've developed an agi script a while ago to use google speech recognition and by then I've used http://legroom.net/files/software/convtoflac.sh to convert files from wav to flac. You can the use the command: */usr/local/bin/convtoflac.sh -o /var/lib/asterisk/sounds/myfile.wav* It will then create create a flac file in the same directory as the source file. I hope it helps. regards Lobito On 1/4/2012 5:51 AM, Bruce B wrote: And with recent version 14.3.2 I get: /usr/local/bin/sox FAIL formats: no handler for file extension `flac' -- speech-recog.agi: /usr/local/bin/sox failed: 512 -- SIP/-002eAGI Script speech-recog.agi completed, returning 0 Regards, On Wed, Jan 4, 2012 at 12:43 AM, Bruce B bruceb...@gmail.com mailto:bruceb...@gmail.com wrote: Very interesting. I just tried to get it to work but it complains about sox. Probably you used a different version of sox? *PBX-*CLI /usr/bin/sox: invalid option -- -* */usr/bin/sox: invalid option -- n* */usr/bin/sox: invalid option -- o* */usr/bin/sox: -r must be given a positive integer* * -- speech-recog.agi: /usr/bin/sox failed: 512* I am using: *Package sox-12.18.1-1.el5_5.1.i386 * Thanks, On Tue, Jan 3, 2012 at 9:42 PM, Lefteris Zafiris zaf@gmail.com mailto:zaf@gmail.com wrote: Hello, I have written an agi script that uses google voice API for voice recognition. The script records from the current channel untill the pound key (#) is pressed or the timeout (15 seconds) is reached. The recording is send over to google speech recognition service and the returned text string is assigned to a channel variable. More info and dialplan examples can be found in the README file: https://raw.github.com/zaf/asterisk-speech-recog/master/README The script is available here: https://github.com/zaf/asterisk-speech-recog The code is still young and not roughly tested so comments, suggestions and bug reports are more than welcome. Lefteris Zafiris -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[asterisk-users] Speech recognition on simultaneous SIP / PSTN calls
Hi. I'm writing a speech recognition module for Asterisk. I'm having problems with simultaneous SIP and PSTN calls. Sometimes Asterisk crashes in this scenario. I don't have problem with simultaneous calls using PSTN calls only. The implementation is in the file res/res_speech.c Does someone know if the isolation differs in these two protocols? Each function has its own mutex, but the crash still happening. I'm using Asterisk 1.4.21.2 Thank you. -- ___ Allann J. O. Silva I received the fundamentals of my education in school, but that was not enough. My real education, the superstructure, the details, the true architecture, I got out of the public library. For an impoverished child whose family could not afford to buy books, the library was the open door to wonder and achievement, and I can never be sufficiently grateful that I had the wit to charge through that door and make the most of it. (from I. Asimov, 1994) ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- AstriCon 2008 - September 22 - 25 Phoenix, Arizona Register Now: http://www.astricon.net asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[asterisk-users] Speech Recognition Apps
Im thinking of taking another run at www.Tellme.com to set up an open access Pay-As-You-Go SIP gateway for their Speech Recognition services. I tried to do this about a year ago http://www.voip-info.org/wiki/view/Tellme and whilst the initial enthusiasm was good they ended up more or less building the same idea but partnering with Skype. Now that its a year later on hopefully the timing is right and as I have another application that I could build out if this was available want to take another run. In order to build up momentum it would be great to hear from anyone, actually building speech recognition apps with sphinx or who has an existing Asterisk application that could be ported across if this gateway was made available, the more people we get on board the more likely this is to happening. If anyone has anything else to ad please reply or email directly for more confidential matters. Regards, Dean Collins Cognation Pty Ltd [EMAIL PROTECTED] +1-212-203-4357 Ph +1-917-207-3420 Mb +61-2-9016-5642 (Sydney in-dial). ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech recognition or TTS with Asterisk?
On 11/18/05 12:55 John Todd said the following: affordable, which probably means $50 or less I suspect. This would be a native Linux environment for all components. Again, while I have no when, oh when, will folk like these support use downtrodden freebsd folk ? :) -- Regards, /\_/\ All dogs go to heaven. [EMAIL PROTECTED](0 0)http://www.alphaque.com/ +==oOO--(_)--OOo==+ | for a in past present future; do| | for b in clients employers associates relatives neighbours pets; do | | echo The opinions here in no way reflect the opinions of my $a $b. | | done; done | +=+ ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech recognition or TTS with Asterisk?
On Thu, 2005-11-17 at 14:56 -0500, Paul wrote: I provided the link for phpagi. Install it and install festival. Set up an extension going to the weather.php demo. If you want this running real soon contact me offlist about paid services. Otherwise you will be reading and learning. Free help via the list happens while I am taking coffee/donut breaks so patience is a needed virtue. Sadly I found festival easy to set up but really hard to understand what its saying. Unless you are super dedicated to tweaking festival to sound better than speak and spell I would suggest using any number of the paid services. On my webpage I wrote an article talking about how to integrate IBMs TTS engine into asterisk. Its a shell script and a simple macro to call it. Really trivial, but it sounds WAY better. I would also like to add that it appears IBM moved their system, I will see if I can hunt down the new site for their TTS demo ... Their new demo has more voices so that would be a good thing. IBM doesnt sound the best but it is better than festival, and due to the old license anyway, the same cost :) -- Trixter http://www.0xdecafbad.com Bret McDanel UK +44 870 340 4605 Germany +49 801 777 555 3402 US +1 360 207 0479 or +1 516 687 5200 FreeWorldDialup: 635378 http://www.sacaug.org/ Sacramento Asterisk Users Group signature.asc Description: This is a digitally signed message part ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech recognition or TTS with Asterisk?
trixter aka Bret McDanel wrote: On Thu, 2005-11-17 at 14:56 -0500, Paul wrote: I provided the link for phpagi. Install it and install festival. Set up an extension going to the weather.php demo. If you want this running real soon contact me offlist about paid services. Otherwise you will be reading and learning. Free help via the list happens while I am taking coffee/donut breaks so patience is a needed virtue. Sadly I found festival easy to set up but really hard to understand what its saying. Unless you are super dedicated to tweaking festival to sound better than speak and spell I would suggest using any number of the paid services. On my webpage I wrote an article talking about how to integrate IBMs TTS engine into asterisk. Its a shell script and a simple macro to call it. Really trivial, but it sounds WAY better. I would also like to add that it appears IBM moved their system, I will see if I can hunt down the new site for their TTS demo ... Their new demo has more voices so that would be a good thing. IBM doesnt sound the best but it is better than festival, and due to the old license anyway, the same cost :) I use the british male speaker voice. It sounds only slightly better. Nobody listening to my demo has ever accused me of using recorded human speech. They definitely believe it is synthesized. If you follow the links from the festival home page(somewhere at cmu.edu), there are some demos of much better voicing. If you just want a functioning placeholder for whatever you are trying to prototype/develop, festival is a good start. Get some interactive extensions working and look for alternatives when you have time(or money to delegate the task). ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] Speech recognition or TTS with Asterisk?
Hello, I am interested in TTS with Asterisk. Anyone implemented this port? Thanks in adbvance, John B ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech recognition or TTS with Asterisk?
John Brookes wrote: Hello, I am interested in TTS with Asterisk. Anyone implemented this port? Thanks in adbvance, John B Yes. I did it when I installed the phpagi stuff including the weather demo. It worked so I went ahead and strted playing around with it and was able to change things. http://phpagi.sourceforge.net/ I did all this using debian stable (aka sarge) linux. Only thing non-debian is files added to /usr/share/asterisk/agi-bin/ and you might have to Of course I had to edit extensions.conf exten = 17,1,agi(weather.php) exten = 18,1,agi(dtmf.php) exten = 19,1,agi(input.php) exten = 20,1,agi(my_ip.php) I haven't had time to put this on the server running 1.2 rc2 yet. ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech recognition or TTS with Asterisk?
Paul, Can you say more about how I could get started on this? I have been looking at Cepstral for TTS, but any will do. Can this be implemented in Java? JB - Original Message - From: Paul [EMAIL PROTECTED] To: Asterisk Users Mailing List - Non-Commercial Discussion asterisk-users@lists.digium.com Sent: Thursday, November 17, 2005 11:29 AM Subject: Re: [Asterisk-Users] Speech recognition or TTS with Asterisk? John Brookes wrote: Hello, I am interested in TTS with Asterisk. Anyone implemented this port? Thanks in adbvance, John B Yes. I did it when I installed the phpagi stuff including the weather demo. It worked so I went ahead and strted playing around with it and was able to change things. http://phpagi.sourceforge.net/ I did all this using debian stable (aka sarge) linux. Only thing non-debian is files added to /usr/share/asterisk/agi-bin/ and you might have to Of course I had to edit extensions.conf exten = 17,1,agi(weather.php) exten = 18,1,agi(dtmf.php) exten = 19,1,agi(input.php) exten = 20,1,agi(my_ip.php) I haven't had time to put this on the server running 1.2 rc2 yet. ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech recognition or TTS with Asterisk?
John Brookes wrote: Paul, Can you say more about how I could get started on this? I have been looking at Cepstral for TTS, but any will do. Can this be implemented in Java? JB I provided the link for phpagi. Install it and install festival. Set up an extension going to the weather.php demo. If you are running debian 3.1 it should be workable. If you are running another distro there may be differences. If you want this running real soon contact me offlist about paid services. Otherwise you will be reading and learning. Free help via the list happens while I am taking coffee/donut breaks so patience is a needed virtue. - Original Message - From: Paul [EMAIL PROTECTED] To: Asterisk Users Mailing List - Non-Commercial Discussion asterisk-users@lists.digium.com Sent: Thursday, November 17, 2005 11:29 AM Subject: Re: [Asterisk-Users] Speech recognition or TTS with Asterisk? John Brookes wrote: Hello, I am interested in TTS with Asterisk. Anyone implemented this port? Thanks in adbvance, John B Yes. I did it when I installed the phpagi stuff including the weather demo. It worked so I went ahead and strted playing around with it and was able to change things. http://phpagi.sourceforge.net/ I did all this using debian stable (aka sarge) linux. Only thing non-debian is files added to /usr/share/asterisk/agi-bin/ and you might have to Of course I had to edit extensions.conf exten = 17,1,agi(weather.php) exten = 18,1,agi(dtmf.php) exten = 19,1,agi(input.php) exten = 20,1,agi(my_ip.php) I haven't had time to put this on the server running 1.2 rc2 yet. ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech recognition or TTS with Asterisk?
On Thu, 2005-11-17 at 12:10 -0700, John Brookes wrote: Can this be implemented in Java? sure that can be implemented in Java. Have a look at Asterisk-Java at http://asteriskjava.org. Asterisk-Java is to Java what phpagi is to PHP. =Stefan signature.asc Description: This is a digitally signed message part ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech recognition or TTS with Asterisk?
Hello, I am interested in TTS with Asterisk. Anyone implemented this port? Thanks in adbvance, John B While not being true TTS, there are efforts by LumenVox to incorporate speech recognition into Asterisk. They were at Astricon, and they also were present at IP4IT in the Digium booth on Monday/Tuesday of this week. I have spoken with Gerd Graumann at LumenVox, and he says that they are planning for a Q1 release. While I don't have any written details, there were discussions about making a single-user license very affordable, which probably means $50 or less I suspect. This would be a native Linux environment for all components. Again, while I have no specific details, I believe that Digium is working with them to develop a dialplan application that would allow specific word-matching rules to be easily built. http://www.LumenVox.com/ JT ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
I worked with Intellivoice. They did VAD, voice activated dialing, on the switch. You had to dial a number, speed dial on the cel, to get the reco. It worked with any phone. Their research should that speech recognition was more accurate then DTMF dialing. They were doing voice pattern recognition, that is were you record a couple of names that are kept on the switch/pbx, then a person says the name and the switch thing tries to match the phrase. They could only get about 7 names and they had to be very different phrases. The number recognition, saying 0-9 was done differently, but it was 97% accurate vs. about 80-90ish % for hand dialing. The hard part is that Cell at that time was CDMA so there was a bunch of background noise to do the reco on. The Nynex phone guys used it a bunch. And they did it in South America. In the end the VAD turned out to be easier to do with software in the phone doing the reco because of the lousy quality of the audio the switch was trying to do on. So, what have we learned? Being able to say the number is better then using your, in my case fat, fingers to dial a phone. But you need a good Signal to Noise ratio to recognize them. Also you need to able to recognize natural language numbers. Numbers like 1,2,3,4,... are easy. But most people will say ninety-nine, double oh, Transylvania six five thousand Remember, you can't force train a user. Do reco on 0-9 and be prepared for people saying it does not work. Let me know how you make out with this. Race the tyrant Vanderdecken (soon to have a new and improved website.) -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Robert Rozman Sent: Saturday, July 09, 2005 7:39 AM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [Asterisk-Users] Speech Recognition - Original Message - From: Richard Koch [EMAIL PROTECTED] To: asterisk-users@lists.digium.com Sent: Friday, July 08, 2005 4:38 PM Subject: [Asterisk-Users] Speech Recognition Ed, Check this out: http://turnkey-solution.com/asterisk-sphinx.html That got me up in running in no time. -Rick What are you experiences with recognition accuracy and user acceptance ? Any more info you're willing to share will help out others Regards, Rob. -Original Message- From: Ed Greenberg [mailto:edg at greenberg.org] Sent: Friday, 8 July 2005 9:32 AM To: Dean Collins; Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [Asterisk-Users] Speech Recognition Tell me probably is excessive. I just really need to recognize Yes, No, One, Two, Three and Four. The Sphinx suggestion should help though. /edg ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition
Hi, I'm not sure if DTMF is convenient solution for user that has cellular on his ear Regards, Rob. - Original Message - From: Dean Collins [EMAIL PROTECTED] To: Ed Greenberg [EMAIL PROTECTED]; Asterisk Users Mailing List - Non-Commercial Discussion asterisk-users@lists.digium.com Sent: Friday, July 08, 2005 4:15 PM Subject: RE: [Asterisk-Users] Speech Recognition Ed can I ask you a question, Not trying to influence you one way or the other but why deal with the 'issues' of speech recognition when what you are looking to achieve is easily met with dtmf codes. Dtmf, works, is easy to manage and well established. Speech should only be used when you need to enter complex controls with more than '9' easy options etc. Just a thought. Cheers, Dean -Original Message- From: Ed Greenberg [mailto:[EMAIL PROTECTED] Sent: Friday, 8 July 2005 9:32 AM To: Dean Collins; Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [Asterisk-Users] Speech Recognition Tell me probably is excessive. I just really need to recognize Yes, No, One, Two, Three and Four. The Sphinx suggestion should help though. /edg --On Friday, July 08, 2005 8:27 AM -0400 Dean Collins [EMAIL PROTECTED] wrote: Hi Ed, Did you read the wiki comment on Tellme? Cheers, Dean ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition
- Original Message - From: Richard Koch [EMAIL PROTECTED] To: asterisk-users@lists.digium.com Sent: Friday, July 08, 2005 4:38 PM Subject: [Asterisk-Users] Speech Recognition Ed, Check this out: http://turnkey-solution.com/asterisk-sphinx.html That got me up in running in no time. -Rick What are you experiences with recognition accuracy and user acceptance ? Any more info you're willing to share will help out others Regards, Rob. -Original Message- From: Ed Greenberg [mailto:edg at greenberg.org] Sent: Friday, 8 July 2005 9:32 AM To: Dean Collins; Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [Asterisk-Users] Speech Recognition Tell me probably is excessive. I just really need to recognize Yes, No, One, Two, Three and Four. The Sphinx suggestion should help though. /edg ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
Ed, Please let me know how you make out. I am sort of keeping tract of what asterisk needs for speech. I am not working a project yet, just trying to get a feel for what people need before I start making new stuff. Race the tyrant Vanderdecken Race at code tyrant dot com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ed Greenberg Sent: Friday, July 08, 2005 10:31 AM To: Dean Collins; Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [Asterisk-Users] Speech Recognition --On Friday, July 08, 2005 10:15 AM -0400 Dean Collins [EMAIL PROTECTED] wrote: Ed can I ask you a question, Not trying to influence you one way or the other but why deal with the 'issues' of speech recognition when what you are looking to achieve is easily met with dtmf codes. Believe me... If I can sell your point of view to my client, I won't be doing speech. Unfortunately more IVRs are taking speech and my clients want it for a selling point. We do speech recognition - the competition doesn't. ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] Speech Recognition
I've been asked to integrate some simple speech recognition with an IVR. Is there anything that people are using with Asterisk for this? Where should I start reading? /edg ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
Hi Ed, Did you read the wiki comment on Tellme? Cheers, Dean -Original Message- From: [EMAIL PROTECTED] [mailto:asterisk-users- [EMAIL PROTECTED] On Behalf Of Ed Greenberg Sent: Friday, 8 July 2005 3:51 AM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: [Asterisk-Users] Speech Recognition I've been asked to integrate some simple speech recognition with an IVR. Is there anything that people are using with Asterisk for this? Where should I start reading? /edg ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
On Fri, 2005-07-08 at 08:27 -0400, Dean Collins wrote: Hi Ed, Did you read the wiki comment on Tellme? Cheers, Dean Dont forget sphinx, which I dont know the state of -- Trixter http://www.0xdecafbad.com Bret McDanel UK +44 870 340 4605 Germany +49 801 777 555 3402 US +1 360 207 0479 or +1 516 687 5200 FreeWorldDialup: 635378 signature.asc Description: This is a digitally signed message part ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
Tell me probably is excessive. I just really need to recognize Yes, No, One, Two, Three and Four. The Sphinx suggestion should help though. /edg --On Friday, July 08, 2005 8:27 AM -0400 Dean Collins [EMAIL PROTECTED] wrote: Hi Ed, Did you read the wiki comment on Tellme? Cheers, Dean ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
Ed can I ask you a question, Not trying to influence you one way or the other but why deal with the 'issues' of speech recognition when what you are looking to achieve is easily met with dtmf codes. Dtmf, works, is easy to manage and well established. Speech should only be used when you need to enter complex controls with more than '9' easy options etc. Just a thought. Cheers, Dean -Original Message- From: Ed Greenberg [mailto:[EMAIL PROTECTED] Sent: Friday, 8 July 2005 9:32 AM To: Dean Collins; Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [Asterisk-Users] Speech Recognition Tell me probably is excessive. I just really need to recognize Yes, No, One, Two, Three and Four. The Sphinx suggestion should help though. /edg --On Friday, July 08, 2005 8:27 AM -0400 Dean Collins [EMAIL PROTECTED] wrote: Hi Ed, Did you read the wiki comment on Tellme? Cheers, Dean ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
--On Friday, July 08, 2005 10:15 AM -0400 Dean Collins [EMAIL PROTECTED] wrote: Ed can I ask you a question, Not trying to influence you one way or the other but why deal with the 'issues' of speech recognition when what you are looking to achieve is easily met with dtmf codes. Believe me... If I can sell your point of view to my client, I won't be doing speech. Unfortunately more IVRs are taking speech and my clients want it for a selling point. We do speech recognition - the competition doesn't. ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] Speech Recognition
Ed, Check this out: http://turnkey-solution.com/asterisk-sphinx.html That got me up in running in no time. -Rick -Original Message- From: Ed Greenberg [mailto:edg at greenberg.org] Sent: Friday, 8 July 2005 9:32 AM To: Dean Collins; Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [Asterisk-Users] Speech Recognition Tell me probably is excessive. I just really need to recognize Yes, No, One, Two, Three and Four. The Sphinx suggestion should help though. /edg ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] speech recognition V 2.0
Greetings David, PerlBox would not be usable for the level of service that is needed by Asterisk to be viable Speech. PerlBox is a vocabulary based recognizer, or I as I call it a grunter, where you grunt something and it then does something cute. Grunters depend on you creating a vocabulary list that is different enough in the syllables so that it can tell cookie from kooky. So long as each grunt is different you can get a response from it. But a grunter can't do computer, call my mother-in-law in France using the PSTN connection. It can do Mom. If you talk to a computer the way you talk to a dog you don't get much more then sit, stay and down. The problem and the reason such Reco has never gained support is like all Voice/Speech Activated Dialing engines is that you have to remember all the vocabulary to use it. If you want to call Robert you can't say Bob. Grammar based recognition is the only solution. I know because I have watched several companies' which demanded that they can do it with grunter, which go out of business because the Stupid Customers won't learn how to use the system correctly. Maybe we should refer to the grunter engines as the Neanderthal engines. Not so much an evolutionary step so much as a evolutionary sidetrack experiment that went down a dead end. Grammar reco is not descendent from grunters; it starts further up the tree and is a distinct evolutionary line. As to PerlBox, I give kudos for their efforts for watch they have accomplished and their efforts are to be applauded for its segment. But it will not be able to progress into a useful and widely accepted Asterisk add-on. Sphinx and Festival are good projects. The last I worked with sphinx I was told that it would need modifications to make it more grammar aware, but that was 2 years ago and things may have improved. If not then Sphinx people please let me know when you will add grammars natively or refer me a grammar based engine. Race The Tyrant Vanderdecken -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of David D. Faerman Sent: Tuesday, February 15, 2005 9:56 PM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: [Asterisk-Users] speech recognition V 2.0 hi seraching for info in the chat and in the web i found perlbox to meake speech recognition some one have any experience? any who to to put it to work? any help please thanks David ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] speech recognition V 2.0
Sphinx and Festival are good projects. The last I worked with sphinx I was told that it would need modifications to make it more grammar aware, but that was 2 years ago and things may have improved. If not then Sphinx people please let me know when you will add grammars natively or refer me a grammar based engine. Sphinx and Festival are in fact the current state of the art. you are not likely to find anything better. Sphinx can return a probibility network. You can then attempt to parse paths through the network and use the first path (searching in probibillity order) that parses correctly. You can use a LEX/YACC parser and do well enough. (Get the O'Reilly LEX/YACC book. It's easy to use.) I'm impressed with YACC's performance. I have an application with hundres of grammar rules that runs as fast as UNIX's wc utility. Users _can_ learn the subset of grammer. Remember the game zork or the other text based adventure games? People caught on to the limited subset of English. = Chris Albertson Home: 310-376-1029 [EMAIL PROTECTED] Cell: 310-990-7550 Office: 310-336-5189 [EMAIL PROTECTED] KG6OMK __ Do you Yahoo!? Yahoo! Mail - You care about security. So do we. http://promotions.yahoo.com/new_mail ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] speech recognition V 2.0
- Original Message - From: Race Vanderdecken [EMAIL PROTECTED] To: 'Asterisk Users Mailing List - Non-Commercial Discussion' asterisk-users@lists.digium.com Sent: Wednesday, February 16, 2005 6:57 PM Subject: RE: [Asterisk-Users] speech recognition V 2.0 Greetings David, PerlBox would not be usable for the level of service that is needed by Asterisk to be viable Speech. PerlBox is a vocabulary based recognizer, or I as I call it a grunter, where you grunt something and it then does something cute. Grunters depend on you creating a vocabulary list that is different enough in the syllables so that it can tell cookie from kooky. So long as each grunt is different you can get a response from it. But a grunter can't do computer, call my mother-in-law in France using the PSTN connection. It can do Mom. If you talk to a computer the way you talk to a dog you don't get much more then sit, stay and down. The problem and the reason such Reco has never gained support is like all Voice/Speech Activated Dialing engines is that you have to remember all the vocabulary to use it. If you want to call Robert you can't say Bob. Grammar based recognition is the only solution. I know because I have watched several companies' which demanded that they can do it with grunter, which go out of business because the Stupid Customers won't learn how to use the system correctly. Maybe we should refer to the grunter engines as the Neanderthal engines. Not so much an evolutionary step so much as a evolutionary sidetrack experiment that went down a dead end. Grammar reco is not descendent from grunters; it starts further up the tree and is a distinct evolutionary line. As to PerlBox, I give kudos for their efforts for watch they have accomplished and their efforts are to be applauded for its segment. But it will not be able to progress into a useful and widely accepted Asterisk add-on. Sphinx and Festival are good projects. The last I worked with sphinx I was told that it would need modifications to make it more grammar aware, but that was 2 years ago and things may have improved. If not then Sphinx people please let me know when you will add grammars natively or refer me a grammar based engine. Hi, there are some initial attempts to connect Misterhouse and Sphinx4 to dinamically transfer grammar that is to be recognized. There could be some colaboration here... Regards, Rob. ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] speech recognition V 2.0
thanks to all for the responce the idea is to use a grunt system someone say david and asterisk tranfer to me i dont care if some stupid cannt say david i will put the option to press the number if no valid option is selected in the speech recognition please if someone can say to me how to put perlbox whit asterisk i will thanks thanks! david - Original Message - From: Robert Rozman [EMAIL PROTECTED] To: Asterisk Users Mailing List - Non-Commercial Discussion asterisk-users@lists.digium.com Sent: Wednesday, February 16, 2005 5:15 PM Subject: Re: [Asterisk-Users] speech recognition V 2.0 - Original Message - From: Race Vanderdecken [EMAIL PROTECTED] To: 'Asterisk Users Mailing List - Non-Commercial Discussion' asterisk-users@lists.digium.com Sent: Wednesday, February 16, 2005 6:57 PM Subject: RE: [Asterisk-Users] speech recognition V 2.0 Greetings David, PerlBox would not be usable for the level of service that is needed by Asterisk to be viable Speech. PerlBox is a vocabulary based recognizer, or I as I call it a grunter, where you grunt something and it then does something cute. Grunters depend on you creating a vocabulary list that is different enough in the syllables so that it can tell cookie from kooky. So long as each grunt is different you can get a response from it. But a grunter can't do computer, call my mother-in-law in France using the PSTN connection. It can do Mom. If you talk to a computer the way you talk to a dog you don't get much more then sit, stay and down. The problem and the reason such Reco has never gained support is like all Voice/Speech Activated Dialing engines is that you have to remember all the vocabulary to use it. If you want to call Robert you can't say Bob. Grammar based recognition is the only solution. I know because I have watched several companies' which demanded that they can do it with grunter, which go out of business because the Stupid Customers won't learn how to use the system correctly. Maybe we should refer to the grunter engines as the Neanderthal engines. Not so much an evolutionary step so much as a evolutionary sidetrack experiment that went down a dead end. Grammar reco is not descendent from grunters; it starts further up the tree and is a distinct evolutionary line. As to PerlBox, I give kudos for their efforts for watch they have accomplished and their efforts are to be applauded for its segment. But it will not be able to progress into a useful and widely accepted Asterisk add-on. Sphinx and Festival are good projects. The last I worked with sphinx I was told that it would need modifications to make it more grammar aware, but that was 2 years ago and things may have improved. If not then Sphinx people please let me know when you will add grammars natively or refer me a grammar based engine. Hi, there are some initial attempts to connect Misterhouse and Sphinx4 to dinamically transfer grammar that is to be recognized. There could be some colaboration here... Regards, Rob. ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] speech recognition V 2.0
hi seraching for info in the chat and in the web i found perlbox to meake speech recognition some one have any experience? any who to to put it to work? any help please thanks David ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] speech recognition
hi i am looking for some info for speech recognition for example when someone call to my house asterisk ask for who hi want to call and he say the name david or susan (wife) or daniela etc... thanks David ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] speech recognition
David D. Faerman wrote: hi i am looking for some info for speech recognition for example when someone call to my house asterisk ask for who hi want to call and he say the name david or susan (wife) or daniela etc... And the wife asks Who's Daniela? ;-) -- _/_/_/_/ _/ _/ _/_/ _/ _/ _/ _/_/_/_/ _/ _/_/ _/ _/ _/ _/_/_/_/ _/ _/ _/ Bill Maidment Maidment Enterprises Pty Ltd Unless you are named Alfred E. Newman, you may read only the odd numbered words (every other word beginning with the first) of the message above. If you have violated that, then you hereby owe the sender AU$10 for each even numbered word you have read. Adapted from Stupid Email Disclaimers (see http://www.goldmark.org/jeff/stupid-disclaimers/) ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] speech recognition
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi David D. Faerman wrote: | hi i am looking for some info for speech recognition for example | when someone call to my house asterisk ask for who hi want to call | and he say the name david or susan (wife) or daniela etc... | Why not the easy way ? "Press 1 for Susan", "Press 2 for David", "Press 3 for Sam the Dog", "Press 4 for Nemo the Little Fish", "Press 5 to leave a message", "Press 6 to Hangup". rgds Joo Amaro | thanks David | | | ___ Asterisk-Users | mailing list Asterisk-Users@lists.digium.com | http://lists.digium.com/mailman/listinfo/asterisk-users To | UNSUBSCRIBE or update options visit: | http://lists.digium.com/mailman/listinfo/asterisk-users | | -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFCEJ1SJUm/Bor63CERAvLdAJ9U0nKUzxFy/azVbe/ZgtDQ/WiKCQCgk247 EOJGYXBusZBxL94Pj/Pw/HU= =hVFw -END PGP SIGNATURE- ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] speech recognition
I am not much into speech recognition, but I know that a major company only had success when they simplified the menus so as to only ask simple yes/no-questions in this manner: Do you have problems with your internet connection? (yes = Do you have a black modem?) (no = Do you have problems with your telephone?) In sequence you would be guided through simple yes/no questions. It works like a charm. ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] speech recognition
daniela is affear but shhh - Original Message - From: Bill Maidment [EMAIL PROTECTED] To: Asterisk Users Mailing List - Non-Commercial Discussion asterisk-users@lists.digium.com Sent: Monday, February 14, 2005 8:54 AM Subject: Re: [Asterisk-Users] speech recognition David D. Faerman wrote: hi i am looking for some info for speech recognition for example when someone call to my house asterisk ask for who hi want to call and he say the name david or susan (wife) or daniela etc... And the wife asks Who's Daniela? ;-) -- _/_/_/_/ _/ _/ _/_/ _/ _/ _/ _/_/_/_/ _/ _/_/ _/ _/ _/ _/_/_/_/ _/ _/ _/ Bill Maidment Maidment Enterprises Pty Ltd Unless you are named Alfred E. Newman, you may read only the odd numbered words (every other word beginning with the first) of the message above. If you have violated that, then you hereby owe the sender AU$10 for each even numbered word you have read. Adapted from Stupid Email Disclaimers (see http://www.goldmark.org/jeff/stupid-disclaimers/) ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition
Hi Dean, What relevance has that to what we were discussing? We were talking about free form speech to text. That is a world apart from a voice activated IVR. Besides that, I have never found a voice activated IVR in English that gets better than about 30% accuracy on a fairly limited decision. A slight divergence from the typical 98% they claim. In contrast, I have seen very good accuracy for Cantonese and Mandarin, which have been less intensively developed. Regards, Steve dean collins wrote: Disagree with you Matt. Check out www.angel.com If anyone wants some contacts over there email me. I'm sure they would be happy to set up on API for utilizing their services in conjunction with asterisk. Cheers, Dean -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Matt Klein Sent: Saturday, February 12, 2005 11:44 PM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [Asterisk-Users] Speech Recognition Agreed, Steve. Iq, Maybe it is for your voice, but speech to text is a long ways away from being as advanced as you think it is. Check out dragon speek, and see what it takes to train a voice... -m On Sun, 13 Feb 2005, Steve Underwood wrote: Iqbal wrote: Hi I dont know jack about speech recognition, however since this topic came up anyonw know how spinvox do speech ercognition, in fact its so good it converst the speech to text and sends the voicemail as a SMS, I think a awesome addone to the sms module in asterisk. If it works really well, there is probably a human operator involved. A number of systems that try to look automated actually rely on human operators. Regards, Steve ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
Steve then you have had your head up your arse for a number of years. Nuance was delivering 90% in 1999 and I have a number of happy customers to prove it. You also obviously didn't look at either the Nuance or angel sites because both of them offer free form speech to text capabilities. One of the first customers I had in Australia for Nuance was ordering of stock for Revlon cosmetics using a speech to an automated ordering system using their antiquated stock database. Dean -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Steve Underwood Sent: Sunday, February 13, 2005 6:52 AM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [Asterisk-Users] Speech Recognition Hi Dean, What relevance has that to what we were discussing? We were talking about free form speech to text. That is a world apart from a voice activated IVR. Besides that, I have never found a voice activated IVR in English that gets better than about 30% accuracy on a fairly limited decision. A slight divergence from the typical 98% they claim. In contrast, I have seen very good accuracy for Cantonese and Mandarin, which have been less intensively developed. Regards, Steve dean collins wrote: Disagree with you Matt. Check out www.angel.com If anyone wants some contacts over there email me. I'm sure they would be happy to set up on API for utilizing their services in conjunction with asterisk. Cheers, Dean -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Matt Klein Sent: Saturday, February 12, 2005 11:44 PM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [Asterisk-Users] Speech Recognition Agreed, Steve. Iq, Maybe it is for your voice, but speech to text is a long ways away from being as advanced as you think it is. Check out dragon speek, and see what it takes to train a voice... -m On Sun, 13 Feb 2005, Steve Underwood wrote: Iqbal wrote: Hi I dont know jack about speech recognition, however since this topic came up anyonw know how spinvox do speech ercognition, in fact its so good it converst the speech to text and sends the voicemail as a SMS, I think a awesome addone to the sms module in asterisk. If it works really well, there is probably a human operator involved. A number of systems that try to look automated actually rely on human operators. Regards, Steve ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition
Hi Dean, You seem to have had your head up the supplier's arse for a number of years. :-) I last tried a Nuance demo system in about 2002, and found it useless. Speechworks (now scansoft) was rather better, but still useless for English. I'm British. Trying the British system gave poor results. Trying the US system seldom gave the right answer. Speechwork's Chinese (Cantonese and Mandarin) was pretty good, though. I've never seen Nuance offer free form speech to text, and I can't see Angel or Nuance's sites claiming that. They offer free form IVR input within a limited domain, which is something quite different - the set of possible outcomes is so much smaller. The best free form speech to text systems still require considerable user specific training to achieve reasonable accuracy. Some people eventually get good results, while others never do. Maybe some people just talk in a much more consistent way. Regards, Steve dean collins wrote: Steve then you have had your head up your arse for a number of years. Nuance was delivering 90% in 1999 and I have a number of happy customers to prove it. You also obviously didn't look at either the Nuance or angel sites because both of them offer free form speech to text capabilities. One of the first customers I had in Australia for Nuance was ordering of stock for Revlon cosmetics using a speech to an automated ordering system using their antiquated stock database. Dean ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
The limited domain reference is obsolete, Telstra have a 2 million record database (yeh I know it's a lot smaller when you dice it phonetically but it's still big enough). -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Steve Underwood Sent: Sunday, February 13, 2005 9:57 AM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [Asterisk-Users] Speech Recognition Hi Dean, You seem to have had your head up the supplier's arse for a number of years. :-) I last tried a Nuance demo system in about 2002, and found it useless. Speechworks (now scansoft) was rather better, but still useless for English. I'm British. Trying the British system gave poor results. Trying the US system seldom gave the right answer. Speechwork's Chinese (Cantonese and Mandarin) was pretty good, though. I've never seen Nuance offer free form speech to text, and I can't see Angel or Nuance's sites claiming that. They offer free form IVR input within a limited domain, which is something quite different - the set of possible outcomes is so much smaller. The best free form speech to text systems still require considerable user specific training to achieve reasonable accuracy. Some people eventually get good results, while others never do. Maybe some people just talk in a much more consistent way. Regards, Steve dean collins wrote: Steve then you have had your head up your arse for a number of years. Nuance was delivering 90% in 1999 and I have a number of happy customers to prove it. You also obviously didn't look at either the Nuance or angel sites because both of them offer free form speech to text capabilities. One of the first customers I had in Australia for Nuance was ordering of stock for Revlon cosmetics using a speech to an automated ordering system using their antiquated stock database. Dean ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
On Mon, February 14, 2005 2:18, dean collins said: The limited domain reference is obsolete, Telstra have a 2 million record database (yeh I know it's a lot smaller when you dice it phonetically but it's still big enough). Maybe it's just me, but I found their database very hit and miss, not to mentioned biased towards their own services, for things such as internet... -- Best regards, Duane http://www.cacert.org - Free Security Certificates http://www.nodedb.com - Think globally, network locally http://www.sydneywireless.com - Telecommunications Freedom http://happysnapper.com.au - Sell your photos over the net! http://e164.org - Using Enum.164 to interconnect asterisk servers In the long run the pessimist may be proved right, but the optimist has a better time on the trip. ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
Oh yeh, their database admins have been playing funny games with the rules. It's been demonstrated on more than a few 'key words' -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Duane Sent: Sunday, February 13, 2005 10:25 AM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [Asterisk-Users] Speech Recognition On Mon, February 14, 2005 2:18, dean collins said: The limited domain reference is obsolete, Telstra have a 2 million record database (yeh I know it's a lot smaller when you dice it phonetically but it's still big enough). Maybe it's just me, but I found their database very hit and miss, not to mentioned biased towards their own services, for things such as internet... -- Best regards, Duane http://www.cacert.org - Free Security Certificates http://www.nodedb.com - Think globally, network locally http://www.sydneywireless.com - Telecommunications Freedom http://happysnapper.com.au - Sell your photos over the net! http://e164.org - Using Enum.164 to interconnect asterisk servers In the long run the pessimist may be proved right, but the optimist has a better time on the trip. ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
Hi I dont know jack about speech recognition, however since this topic came up anyonw know how spinvox do speech ercognition, in fact its so good it converst the speech to text and sends the voicemail as a SMS, I think a awesome addone to the sms module in asterisk. Iqbal On 2/12/2005, Race Vanderdecken [EMAIL PROTECTED] wrote: Ahem, Being one who has programmed, consulted and argued to points beyond violence about the subjects of your first paragraph, I shall now expound. Expounding begins: I worked on several projects with a company named Intellivoice that did so called voice dialing, voice activated dialing, VAD, as a bread and butter product in the PSTN/T-1 world. The product was good at about 3-5 recognitions so long as they were distinct enough that your well trained dog could understand them as different commands. I was first hand witness to many sales and customer meetings, I rode in the car of the inventor and ate lunch with the VAD developers and beat them often with questions about how they did it and why it did not work. Personally, I have a Mid-Western trained Mid-Atlantic accent, i.e. no accent to speak of, so speech and voice recognition engines like me. I am even tempered and have been working in telecommunications, 22 wpm Morse code, to Tech Plus, to before NETBIOS, SNB, and 256K twisted pair Ethernet on 9DB, through voice and right back into VoIP before it was an acronym. I have been to college to study communications. I have an ear for dialects and can place most people in 100 mile range within their State. I coded the Persona project. I have pushed Sphinx down Festivals throat, and I have worked with Dave. I was working to create voice X/HTML/XML browser before they were committees. I am pushed speech and voice and dictation since I got my hands on a computer. I love speech recognition and generation, period. So, when I say that you are out of your mind if you think you can get VAD or SAD to work across the wire if there is an analog device in the path you should take heed. VAD on cel-phones works now because the reco is in the phone, for the most part. VAD over the analog wires can be done but is of no use to anyone unless they like to scream at the phone from time to time. By the way screaming at voice-reco engines only makes the angry. So angry in fact that they will either repeatedly ask you to please say the name again until you calm down or they will deliberately misdial the number for you. Machines just don't like to be yelled at, ask Woody Allen about the time beat up his television and the elevator incident. If you are Digital from speaker to reco then you have a chance. If you are G.711 all the way you have a chance. And by chance I mean if you use grammar based recognition and have a caller with an IQ greater then the first two digit of their Area code. Zhong's thesis work is interesting and I will state he is on the right track and I enjoin him to continue his work. Neural works are the right path, but he needs to re-read Strousstrup on objects in Tries. But short utterances don't work in communication models; see trying to communicate with teenage son and HDLC. My Expounding ends. Yes, what you want can be done, and it would be easy, no, I say trivial to accomplish. Follow the dynamic context example. Listen for reco, then create a dynamic context, then forward to that dynamic context. Easy peasy. Please contact me, [EMAIL PROTECTED], if you would like to create a project to do this. I would love to see VAD on VoIP come to fruition in my lifetime. Race The Tyrant Vanderdecken -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Robert Rozman Sent: Saturday, January 29, 2005 12:16 PM To: Jon Radon; Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [Asterisk-Users] Speech Recognition Hi, probably I won't be much of help, but I'm also looking for speech recognition solution. But we're actually looking at two problems: - one would be so called voice dialing (similar to celular phones) - one records its own spoken names and speaks them after to call certain person - this problem is much easier to solve. Recently I have found interesting project that could be easily integrated for such functionality (http://www.princeton.edu/~lzhong/DNN.html). I'd like to start doing this but don't know much about Asterisk and its eagi interface to get sound out of it. I guess some with more insight could easily integrate this code. This solution could be probably used for simple 1 word recognition tasks (like speak name for outgoing call, or maybe say sales to get sales department - but as said this is speaker dependent solution. - using speaker independent solutions for other stuff. I guess that Sphinx is at the moment most serious candidate. There is already some work on connecting speech recognition to MH and I'm sure that guys will help with other uses. What would be most desirable from Asterisk community
Re: [Asterisk-Users] Speech Recognition
Iqbal wrote: Hi I dont know jack about speech recognition, however since this topic came up anyonw know how spinvox do speech ercognition, in fact its so good it converst the speech to text and sends the voicemail as a SMS, I think a awesome addone to the sms module in asterisk. If it works really well, there is probably a human operator involved. A number of systems that try to look automated actually rely on human operators. Regards, Steve ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition
Does anyone know of a speech recognition module (like say yes or no, or numbers) I guess due to the complexity of speech recognition it might just be found in commercial applications or am I wrong like always? What's wrong with the old and non-fancy IVR? Voice recognition menus only piss people off. If you're setting up a call center where you want as many as possible of the customers to ABANDON their calls, go on... roy ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition
On Feb 12, 2005, at 17:58, Roy Sigurd Karlsbakk wrote: Does anyone know of a speech recognition module (like say yes or no, or numbers) I guess due to the complexity of speech recognition it might just be found in commercial applications or am I wrong like always? What's wrong with the old and non-fancy IVR? Voice recognition menus only piss people off. If you're setting up a call center where you want as many as possible of the customers to ABANDON their calls, go on... How true that is... faced with customer-unfriendly service like that (especially when they don't offer a choice to get a human at all) I start hitting keys like 0 or # or * until something happens... jens ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition
Agreed, Steve. Iq, Maybe it is for your voice, but speech to text is a long ways away from being as advanced as you think it is. Check out dragon speek, and see what it takes to train a voice... -m On Sun, 13 Feb 2005, Steve Underwood wrote: Iqbal wrote: Hi I dont know jack about speech recognition, however since this topic came up anyonw know how spinvox do speech ercognition, in fact its so good it converst the speech to text and sends the voicemail as a SMS, I think a awesome addone to the sms module in asterisk. If it works really well, there is probably a human operator involved. A number of systems that try to look automated actually rely on human operators. Regards, Steve ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
Disagree with you Matt. Check out www.angel.com If anyone wants some contacts over there email me. I'm sure they would be happy to set up on API for utilizing their services in conjunction with asterisk. Cheers, Dean -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Matt Klein Sent: Saturday, February 12, 2005 11:44 PM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [Asterisk-Users] Speech Recognition Agreed, Steve. Iq, Maybe it is for your voice, but speech to text is a long ways away from being as advanced as you think it is. Check out dragon speek, and see what it takes to train a voice... -m On Sun, 13 Feb 2005, Steve Underwood wrote: Iqbal wrote: Hi I dont know jack about speech recognition, however since this topic came up anyonw know how spinvox do speech ercognition, in fact its so good it converst the speech to text and sends the voicemail as a SMS, I think a awesome addone to the sms module in asterisk. If it works really well, there is probably a human operator involved. A number of systems that try to look automated actually rely on human operators. Regards, Steve ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
dean collins wrote on Saturday, 12 February 2005 9:16 PM: Check out www.angel.com For that matter, check out Tellme. 1-800-555-TELL Speaker-independent automatic speech recognition, when implemented properly, is VERY good right now. However, good ASR is usually fairly expensive. Do not confuse desktop speech recognition applications like Dragon Dictate and Via Voice with telco-grade ASR engines like Nuance, SpeechPearl/Speechworks, Loquendo, etc. Tellme has a developer platform that you can use to experiment with VoiceXML, TTS, and ASR. You create the voice applications on their website, and can access them via the PSTN or SIP. Check out: http://studio.tellme.com/ I, for one, would love to have the ability to use ASR engines with Asterisk. I think a good start would be a Sphinx/Asterisk integration project. Sincerely, Trevor Hammonds ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
Trevor, I used to sell Nuance when I worked for Fujitsu. When I first came across Angel.com about 12 months ago I knew this was the right way to approach NLVR. The Nuance costs are unrealistically astronomical mainly due to the 'high touch' consulting fees that are imposed on this kind of rollout. Angel.com being web based and delivered is able to deliver 90% of the solution for 20% of the costs can only be a good thing. Cheers, Dean -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Trevor G. Hammonds Sent: Sunday, February 13, 2005 1:46 AM To: 'Asterisk Users Mailing List - Non-Commercial Discussion' Subject: RE: [Asterisk-Users] Speech Recognition dean collins wrote on Saturday, 12 February 2005 9:16 PM: Check out www.angel.com For that matter, check out Tellme. 1-800-555-TELL Speaker-independent automatic speech recognition, when implemented properly, is VERY good right now. However, good ASR is usually fairly expensive. Do not confuse desktop speech recognition applications like Dragon Dictate and Via Voice with telco-grade ASR engines like Nuance, SpeechPearl/Speechworks, Loquendo, etc. Tellme has a developer platform that you can use to experiment with VoiceXML, TTS, and ASR. You create the voice applications on their website, and can access them via the PSTN or SIP. Check out: http://studio.tellme.com/ I, for one, would love to have the ability to use ASR engines with Asterisk. I think a good start would be a Sphinx/Asterisk integration project. Sincerely, Trevor Hammonds ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition
Ahem, Being one who has programmed, consulted and argued to points beyond violence about the subjects of your first paragraph, I shall now expound. Expounding begins: I worked on several projects with a company named Intellivoice that did so called voice dialing, voice activated dialing, VAD, as a bread and butter product in the PSTN/T-1 world. The product was good at about 3-5 recognitions so long as they were distinct enough that your well trained dog could understand them as different commands. I was first hand witness to many sales and customer meetings, I rode in the car of the inventor and ate lunch with the VAD developers and beat them often with questions about how they did it and why it did not work. Personally, I have a Mid-Western trained Mid-Atlantic accent, i.e. no accent to speak of, so speech and voice recognition engines like me. I am even tempered and have been working in telecommunications, 22 wpm Morse code, to Tech Plus, to before NETBIOS, SNB, and 256K twisted pair Ethernet on 9DB, through voice and right back into VoIP before it was an acronym. I have been to college to study communications. I have an ear for dialects and can place most people in 100 mile range within their State. I coded the Persona project. I have pushed Sphinx down Festivals throat, and I have worked with Dave. I was working to create voice X/HTML/XML browser before they were committees. I am pushed speech and voice and dictation since I got my hands on a computer. I love speech recognition and generation, period. So, when I say that you are out of your mind if you think you can get VAD or SAD to work across the wire if there is an analog device in the path you should take heed. VAD on cel-phones works now because the reco is in the phone, for the most part. VAD over the analog wires can be done but is of no use to anyone unless they like to scream at the phone from time to time. By the way screaming at voice-reco engines only makes the angry. So angry in fact that they will either repeatedly ask you to please say the name again until you calm down or they will deliberately misdial the number for you. Machines just don't like to be yelled at, ask Woody Allen about the time beat up his television and the elevator incident. If you are Digital from speaker to reco then you have a chance. If you are G.711 all the way you have a chance. And by chance I mean if you use grammar based recognition and have a caller with an IQ greater then the first two digit of their Area code. Zhong's thesis work is interesting and I will state he is on the right track and I enjoin him to continue his work. Neural works are the right path, but he needs to re-read Strousstrup on objects in Tries. But short utterances don't work in communication models; see trying to communicate with teenage son and HDLC. My Expounding ends. Yes, what you want can be done, and it would be easy, no, I say trivial to accomplish. Follow the dynamic context example. Listen for reco, then create a dynamic context, then forward to that dynamic context. Easy peasy. Please contact me, [EMAIL PROTECTED], if you would like to create a project to do this. I would love to see VAD on VoIP come to fruition in my lifetime. Race The Tyrant Vanderdecken -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Robert Rozman Sent: Saturday, January 29, 2005 12:16 PM To: Jon Radon; Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [Asterisk-Users] Speech Recognition Hi, probably I won't be much of help, but I'm also looking for speech recognition solution. But we're actually looking at two problems: - one would be so called voice dialing (similar to celular phones) - one records its own spoken names and speaks them after to call certain person - this problem is much easier to solve. Recently I have found interesting project that could be easily integrated for such functionality (http://www.princeton.edu/~lzhong/DNN.html). I'd like to start doing this but don't know much about Asterisk and its eagi interface to get sound out of it. I guess some with more insight could easily integrate this code. This solution could be probably used for simple 1 word recognition tasks (like speak name for outgoing call, or maybe say sales to get sales department - but as said this is speaker dependent solution. - using speaker independent solutions for other stuff. I guess that Sphinx is at the moment most serious candidate. There is already some work on connecting speech recognition to MH and I'm sure that guys will help with other uses. What would be most desirable from Asterisk community is some skeleton code for eagi interface... I also have question: does eagi based recognition take place in parallel to other dialplan activities (like dtmf recognition, actions, etc...) ? Regards, Rob. ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com
Re: [Asterisk-Users] Speech Recognition
Hi! Does anyone know of a speech recognition module (like say yes or no, or numbers) I guess due to the complexity of speech recognition it might just be found in commercial applications or am I wrong like always? Search for sphinx. Cheers, Philipp ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition
I'm not sure searching for Sphinx will do him much good. There's really nothing concrete that I've seen. On Sat, 29 Jan 2005 11:40:32 +0100, Philipp von Klitzing Search for sphinx. Cheers, Philipp -- Is it something someone said, was it something someone said? ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition
Hi, probably I won't be much of help, but I'm also looking for speech recognition solution. But we're actually looking at two problems: - one would be so called voice dialing (similar to celular phones) - one records its own spoken names and speaks them after to call certain person - this problem is much easier to solve. Recently I have found interesting project that could be easily integrated for such functionality (http://www.princeton.edu/~lzhong/DNN.html). I'd like to start doing this but don't know much about Asterisk and its eagi interface to get sound out of it. I guess some with more insight could easily integrate this code. This solution could be probably used for simple 1 word recognition tasks (like speak name for outgoing call, or maybe say sales to get sales department - but as said this is speaker dependent solution. - using speaker independent solutions for other stuff. I guess that Sphinx is at the moment most serious candidate. There is already some work on connecting speech recognition to MH and I'm sure that guys will help with other uses. What would be most desirable from Asterisk community is some skeleton code for eagi interface... I also have question: does eagi based recognition take place in parallel to other dialplan activities (like dtmf recognition, actions, etc...) ? Regards, Rob. ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] Speech Recognition
Does anyone know of a speech recognition module (like say yes or no, or numbers) I guess due to the complexity of speech recognition it might just be found in commercial applications or am I wrong like always? ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition
Here is an open source, speech reco initiative, not sure if this is intended for web applications or what http://freespeech.sourceforge.net/FreeSpeech/html/ Cory Andrews Senior Partner VOIPSupply.com + 800.398.VOIP X22 [EMAIL PROTECTED] Manjit Riat wrote: Does anyone know of a speech recognition module (like say yes or no, or numbers) I guess due to the complexity of speech recognition it might just be found in commercial applications or am I wrong like always? ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] Speech Recognition and Asterisk
All; Since I have interest in providing the capability for callers to speak the department, person or number they wish to call, as well as other IVR scenarios, I have been reviewing much of this lists email archives and searching the web for open source voice recognition that will work with the Asterisk PBX. What I am trying to determine, is what will it take to get it working on Asterisk? How much effort and cost? So far I have uncovered references to the following: 1) VoiceXML standards, and forums 2) OpenVXI - which supports VoiceXML, simulated speech, telephony 3) PublicVoiceXML 4) Sphinx - a Carnegie Mellon University Speech recognition project funded by DARPA From what I can tell, I feel I am uncovering the tip of the ice berg and this may not be trivial. But it seems that the Voice recognition application, once developed, would have to be linked via an AGI to the asterisk dial plan. Has anyone gotten Voice recognition working with Asterisk? Last I saw, a few were attempting to apply Sphinx back in the December and April time frame. Any shared successes, progress or direction on Sphinx or any other VR app would be appreciated before I start down this road. Thanks, Mike Meyer ___ Asterisk-Users mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition and Asterisk
On Fri, 2004-08-27 at 13:26, Mike Meyer wrote: Mike, I have been mulling similar ideas for some time. I've turned up the same projects you have. From what I can tell, I feel I am uncovering the tip of the ice berg and this may not be trivial. I've pretty much got the same feeling based on my research. I don't think this is a trivial problem to solve. But it seems that the Voice recognition application, once developed, would have to be linked via an AGI to the asterisk dial plan. I think this is accurate. You would use either AGI or a native Asterisk module to connect the pieces. Has anyone gotten Voice recognition working with Asterisk? Last I saw, a few were attempting to apply Sphinx back in the December and April time frame. Any shared successes, progress or direction on Sphinx or any other VR app would be appreciated before I start down this road. I have nothing working at this point, just some ideas and a plan to explore further. As soon as I have free time I'm hoping to explore some of my ideas on my * system. I certainly would be happy to hear others' experiences on this mail list. -joe -- Innovation Software Group, LLC - http://www.innovationsw.com Custom Internet and Computer Solutions Linux, UNIX, Java Training ___ Asterisk-Users mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Speech Recognition and Asterisk
I dumped about 2 weeks of my life into doing just batch speech-to-text using Sphinx2. After doing all sorts of custom configurations to the Sphinx batch run-time parameters and using a very limited vocabulary I was able to recognize about 95% of the phrases that were uttered in 4000 random snippets of conversations that I recorded from Asterisk. I was never able to get real-time conversion working in any reliable form over phone quality audio, and due to the processor and memory requirements it would be rather limiting to try using it on a busy IVR system. Sphinx runs best the more RAM it has(suggested minimum 256MB of RAM) and it is very much a processor hog. I would be very interested to hear if anyone has any experiences with IBM's viavoice product in a real-time capacity(even though I understand it is quite expensive for a multi-stream license). In the end Sphinx2 worked well enough for what I needed to do(batch processing of phrases in a limited vocabulary), but nowhere near well enough to try using it real-time in any way. Sphinx4 promises to be much better at conversion, but it is very much still beta at this time. Hope this helps. MATT--- -Original Message- From: Mike Meyer [mailto:[EMAIL PROTECTED] Sent: Friday, August 27, 2004 1:27 PM To: Asterisk Users Group Subject: [Asterisk-Users] Speech Recognition and Asterisk All; Since I have interest in providing the capability for callers to speak the department, person or number they wish to call, as well as other IVR scenarios, I have been reviewing much of this lists email archives and searching the web for open source voice recognition that will work with the Asterisk PBX. What I am trying to determine, is what will it take to get it working on Asterisk? How much effort and cost? So far I have uncovered references to the following: 1) VoiceXML standards, and forums 2) OpenVXI - which supports VoiceXML, simulated speech, telephony 3) PublicVoiceXML 4) Sphinx - a Carnegie Mellon University Speech recognition project funded by DARPA From what I can tell, I feel I am uncovering the tip of the ice berg and this may not be trivial. But it seems that the Voice recognition application, once developed, would have to be linked via an AGI to the asterisk dial plan. Has anyone gotten Voice recognition working with Asterisk? Last I saw, a few were attempting to apply Sphinx back in the December and April time frame. Any shared successes, progress or direction on Sphinx or any other VR app would be appreciated before I start down this road. Thanks, Mike Meyer ___ Asterisk-Users mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ Asterisk-Users mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Speech Recognition and Asterisk
On Fri, 2004-08-27 at 12:26, Mike Meyer wrote: All; Since I have interest in providing the capability for callers to speak the department, person or number they wish to call, as well as other IVR scenarios, I have been reviewing much of this lists email archives and searching the web for open source voice recognition that will work with the Asterisk PBX. This is a very hard problem in total, but parts of it are not too difficult. Voice recognition with no training is hard. IT is hard enough to be the very reason it isn't widly deployed now. What is deployed now is usually limited to small vocabularies. Sounds like your request might be able to handle that. For a fairly simple hookup, you could try to write an application for AGI that records a prompt with either a timeout or a silence threshold. Then send that audio off to a application like sphynx with a small vocabulary list to match against. It should return to you a fairly decent idea of what was said. Do remember though that accents might throw it off so be prepared for other options of input. Last I tried, sphynx wasn't too hard to install, but was a pain to use. I would suggest small recordings over EAGI as you can send audio to sphynx in small chunks and with a vocabulary list of expected terms. -- Steven Critchfield [EMAIL PROTECTED] ___ Asterisk-Users mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] speech recognition system
Hi, Well, I'm looking for a speech recognition system. However, we're using asterisk as PBX and we'd like to setup this features. lready done? any suggestions? Let me know! dan. ___ Asterisk-Users mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users