Re: [asterisk-users] Speech Rec on Voicemail
Ron Joffe wrote: On Friday 24 August 2007 12:37, Ryan M. Colbert wrote: I'd be interesting in pooling resources for this. We've seen the success of Vonage's Visual Voicemail and would like to emulate a similar solution. Please define success, I have a vonage account, and the transcription is very poor at best. I was about to say -- the standards would have to be pretty low to call the Vonage Visual Voicemail a success. I will give them points for daring to try, though. -Stephen- ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech Rec on Voicemail
Stephen Bosch wrote: Ryan M. Colbert wrote: I’ve had requests to processes incoming voicemails with voice recognition routine and add the output text to the body of the email message from * with the attached .wav file. Has anyone implemented this type of feature and willing to share some notes? I seem to recall that Lt. Cmdr Jordi Laforge did some stuff like this not too long ago. I get requests like this all the time -- but the technology is very far from being there. -Stephen- Your best bet at this point is to have an assistant or some other 3rd party transcribe the VM for you. I know of several companies that do this for recordings in various industries. Thanks, Steve ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech Rec on Voicemail
Just for starter, look at CallWave, and Jott. -E ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech Rec on Voicemail
On 8/24/07, EdPimentl [EMAIL PROTECTED] wrote: Just for starter, look at CallWave, and Jott. -E They seem to be commercial :( A quick search in google revealed a page with some compilation of opensource STT engines. http://www.faqs.org/docs/Linux-HOWTO/Speech-Recognition-HOWTO.html Making them process voicemail wav files isn't trivial but shouldn't be quite hard. If you have some progress on this, give us know. Regards, Atis -- Atis Lezdins, IT Responsible of BEST Riga, [EMAIL PROTECTED] ICQ: 142239285 Skype: atis.lezdins Cell Phone: +371 28806004 [Tele2, Latvia] Work phone: +1 800 7502835 [Toll free, USA] ?BEST? - www.BEST.eu.org ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech Rec on Voicemail
I'd be interesting in pooling resources for this. We've seen the success of Vonage's Visual Voicemail and would like to emulate a similar solution. Ryan M. Colbert Director of Information Technology Rissman, Barrett, Hurt, Donahue McLain, P.A. 201 E. Pine Street, Suite 1500 Orlando, FL 32801 (407) 517-3105 - Direct Telephone (407) 839-0120 - Main Office (407) 841-9726 - Fax http://www.rissman.com/ -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of mitcheloc Sent: Friday, August 24, 2007 1:54 AM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [asterisk-users] Speech Rec on Voicemail Nuance offers an SDK to do something similar, I think they say you can only expect between 45-60% accuracy using it though. Total cost is about $6K to $8K for one server license. If there are enough people interested in pooling money I'd be willing to help set up a system to process voicemails and provide the Nuance converted transcript. However, I figure the low accuracy would be the biggest turn off from using Nuance. On 8/23/07, Stephen Bosch [EMAIL PROTECTED] wrote: Ryan M. Colbert wrote: I've had requests to processes incoming voicemails with voice recognition routine and add the output text to the body of the email message from * with the attached .wav file. Has anyone implemented this type of feature and willing to share some notes? I seem to recall that Lt. Cmdr Jordi Laforge did some stuff like this not too long ago. I get requests like this all the time -- but the technology is very far from being there. -Stephen- ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- Mitchel Constantin Snap - A desktop user interface for Asterisk www.snapanumber.com ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- 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-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech Rec on Voicemail
On Friday 24 August 2007 12:37, Ryan M. Colbert wrote: I'd be interesting in pooling resources for this. We've seen the success of Vonage's Visual Voicemail and would like to emulate a similar solution. Please define success, I have a vonage account, and the transcription is very poor at best. Ron ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[asterisk-users] Speech Rec on Voicemail
I've had requests to processes incoming voicemails with voice recognition routine and add the output text to the body of the email message from * with the attached .wav file. Has anyone implemented this type of feature and willing to share some notes? Thanks! Ryan M. Colbert Director of Information Technology Rissman, Barrett, Hurt, Donahue McLain, P.A. 201 E. Pine Street, Suite 1500 Orlando, FL 32801 (407) 517-3105 - Direct Telephone (407) 839-0120 - Main Office (407) 841-9726 - Fax http://www.rissman.com/ ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech Rec on Voicemail
On 8/23/07, Ryan M. Colbert [EMAIL PROTECTED] wrote: I've had requests to processes incoming voicemails with voice recognition routine and add the output text to the body of the email message from * with the attached .wav file. Has anyone implemented this type of feature and willing to share some notes? That would be very interesting to see, if you get it working. Last I checked, though, speech-to-text didn't work very well without a very small language to choose from, far smaller than English. ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech Rec on Voicemail
Ryan M. Colbert wrote: I’ve had requests to processes incoming voicemails with voice recognition routine and add the output text to the body of the email message from * with the attached .wav file. Has anyone implemented this type of feature and willing to share some notes? I seem to recall that Lt. Cmdr Jordi Laforge did some stuff like this not too long ago. I get requests like this all the time -- but the technology is very far from being there. -Stephen- ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Speech Rec on Voicemail
Nuance offers an SDK to do something similar, I think they say you can only expect between 45-60% accuracy using it though. Total cost is about $6K to $8K for one server license. If there are enough people interested in pooling money I'd be willing to help set up a system to process voicemails and provide the Nuance converted transcript. However, I figure the low accuracy would be the biggest turn off from using Nuance. On 8/23/07, Stephen Bosch [EMAIL PROTECTED] wrote: Ryan M. Colbert wrote: I've had requests to processes incoming voicemails with voice recognition routine and add the output text to the body of the email message from * with the attached .wav file. Has anyone implemented this type of feature and willing to share some notes? I seem to recall that Lt. Cmdr Jordi Laforge did some stuff like this not too long ago. I get requests like this all the time -- but the technology is very far from being there. -Stephen- ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- Mitchel Constantin Snap - A desktop user interface for Asterisk www.snapanumber.com ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users