To resurrect the thread: I've been voice-programming with Nuance's Dragon + Natlink <http://qh.antenna.nl/unimacro/> + dragonfly <https://github.com/t4ngo/dragonfly>/caster <https://github.com/synkarius/caster>, a la Tavis Rudd (whose talk Laurent linked to earlier in the thread) since March due to bilateral RSI. The Hands-free Coding blog was invaluable, although I've found that I disagree about its insistence on staying with old Dragon versions. (Dragon 14 seemed to work with native apps well enough.)
Here's an e-mail I'd sent in July with "initial pointers for someone who's trying to get Dragon + voice programming to work": Here's the stupid thing that nobody writes in all caps anywhere: Dragon 15 breaks compatibility with NatLink, or - more specifically - with the parts of NatLink that allowed for continuous command recognition. NatLink maintainers are skeptical that Nuance will fix this in this year's patch, or possibly ever. Hence my use Dragon 14. (On a related note, it took me a while to realize that this is the most up-to-date maintainer of NatLink <http://qh.antenna.nl/unimacro/>.) If you already have Dragon for Mac, you might be interested in trying VoiceCode.io <http://voicecode.io/>? It's a beta that charges $300, but it seems to offer a ready voice programming architecture. I haven't tested it, but it seems to be getting reasonable reviews. Here's a list of useful resources I compiled about a month ago: - *FAQ at Vocola.net* <http://vocola.net/programming-by-voice-FAQ.html> - David Conway’s article series <http://explosionduck.com/wp/tag/voice-programming/> (David is the creater of caster <https://github.com/synkarius/caster>) - Pam Selle’s “Where to learn about voice coding” <http://thewebivore.com/learn-voice-coding/> - Hands-free Coding blog <http://handsfreecoding.org/> - Archived version of the resource list on TheSpanishSite.com <https://web.archive.org/web/20150410201303/http://thespanishsite.com/public_html/org/ergo/programming_by_voice.html> The other excellent list is caster's Voice Index <https://github.com/synkarius/caster/wiki/Voice-Index>. With three months' worth of hindsight, I can't stress enough how useful it was to work with a more-or-less ready system like caster; I was really happy with customizations that dragonfly allowed me to make, but they seem to require a level of Python comfort that goes beyond beginner. (My correspondent, who was a Python beginner, had more success with Vocola instead of dragonfly/caster.) One last thing: having a high-quality headset / microphone with noise cancellation is a must. Starting with a voice programming setup is frustrating; misunderstandings due to under-calibrated voice recognition engine are plentiful enough that you don't need more due to bad audio. Best, Simon On Mon, 18 Sep 2017 at 07:20 Ben Bolker <[email protected]> wrote: > > I don't know if it's useful or not, but there's a blog: > http://handsfreecoding.org/ > <http://handsfreecoding.org/> > I also had some e-mail conversations with colleagues who've had serious > hand problems in the past. They said their experience might be a bit out of > date (~10 years ago), but in particular: > > - being able to use a pointer device (i.e., no keyboard but some > mouse/eyetracking/whatever capability) makes a huge difference > - training the voice recognition system is critical and takes some time > (and a deeper knowledge of the system is required than for typical users). > - software can have very different capabilities on different OSs (e.g. at > that time Dragon Naturally Speaking was usable on Windows, but crippled on > MacOS). > _______________________________________________ > Discuss mailing list > [email protected] > http://lists.software-carpentry.org/listinfo/discuss
_______________________________________________ Discuss mailing list [email protected] http://lists.software-carpentry.org/listinfo/discuss
