The only way to get real speech recognition on Linux is to run Dragon 
NaturallySpeaking version 7, 8 or 9, using WINE. 
Version 9, only Preferred works. I think Standard and Preferred both work for 
versions 7 and 8. Forget Professional, Legal and Medical versions. 
Once you get them installed, they run as fast and as accurately as they do on 
Windows, provided you stick to text recognition. 
Text can be input ONLY into the included DragonPad program or the Notepad 
program that comes with WINE. 

>From there, text can be cut and pasted into other programs using the mouse and 
>keys. 

Follow the directions here:
http://ossri.harvee.org/NaturallySpeakingWine

Start DNS from the desktop icon. Start Notepad from a terminal using the 
command:

wine notepad

If anyone has any difficulty installing, give me a shout on-list using the 
header Speech Recognition --Help or something like that, because I don't read 
all the Studio mailing list entries. 

One tip, if you are new to Dragon NaturallySpeaking ... keep your expectations 
modest. Also, realize that the microphone they ship with the program is a piece 
of crap. If you can't get an accuracy score above 23 for the microphone, throw 
it away and get a new one. (One good thing about this list, I assume you all 
have good microphones.) Good soundcards are also a must. (Another problem you 
probably don't have.) 1GB RAM is essential. A fast processor is good. 

Susan Cragin




-- 
Ubuntu-Studio-users mailing list
[email protected]
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-users

Reply via email to