Dear Lazyweb,
Having broke my elbow last week, I’m experiencing serious trouble typing with only one hand on the computer. I know that there is speech to text software, but I was never really interested in that — until now of course. Is there any good software available on Linux you can recommend?
My feeling is that it’s likely that you will be far better off writing with one hand than using any speech recognition. You might wish to investigate one-hand keyboard layouts, for example one-handed Dvorak; it should ease the pain somewhat.
The only one I know about is pocketsphinx. See http://cmusphinx.sourceforge.net/
Not tested yet.
had the same problem, tried to use dasher for singlehanded mousenavigation textinput, if setting the speed somewhere a bit higher it was easier to use. but gave up after sometime (dont make the mistake to start using your hand too fast again, it will make things worse … lessons learnt)
Get well!
Seconding the recommendations for one-handed Dvorak and for Dasher, depending on how much of a learning curve you want. One-handed Dvorak will likely let you go faster but require more of a learning curve, while Dasher will get you going instantly but not as quickly.
Yep, been wanting decent Linux speech recognition for a while, looked a bit over a year ago and didn’t find much. Maybe it’s specialist enough and involves so much R&D that non-FOSS is the only option at the moment? It’s all a bit odd really, because speech is the natural & least-effort human interface for communicating non-trivial ideas. 10 years ago I assumed all desktop O/Ses or environments in a decade would come with decent speech recognition as standard, just as we now assume TCP/IP, a browser, text editor, some simple games, etc.
CMU Sphinx anyone?
http://cmusphinx.sourceforge.net/wiki/
google speech recognition
dragon naturally speaking works with wine and has faif extensions for programming by voice, ibm had some linux stuff but that was years ago
haven´t you been reading slashdot recently?
http://hardware.slashdot.org/story/11/10/05/1236202/robotic-arm-with-home-brewed-open-source-voice-control
No, there is no good native speech recognition for home usage.
But you can use Dragon Naturally Speaking 10 together with thenerdshow.com/platypus.html
Funny, when I broke my arm a couple years ago I also spend some time looking into speech recognition.
With Free Software only, you can get some “command & control” working (eg. stuff like controlling the music player is quite easy), but there’s no way you can do dictation.
The problem isn’t software, which is available (Pocketsphinx and Julius, both are in Ubuntu and I need to get around uploading them to Debian some day), but the need for voice corpora (recorded speech with the text transcription) to use with those engines. Voxforge (http://www.voxforge.org/) is working on this, but still a lot more corpora is missing for dictation.