(BTW, the phrase "users with no experience in speech" is kinda funny)
Spek 4 u self, I spek gud and spel gudder :]
Seriously though, I would imagine that is just referring to the fact that it doesn't have a nice shiny polished end-user GUI and a zillion and commandline one options with little documentation.
It seems like a decent baseline for someone capable to develop a nice shiny Maemo GUI on top of. This paired with existing flite and the existing text predicition would make for a really nice tool for those with communication impairments a la Stephen Hawking's gadget. Not a huge market segment but an important one.
Coincidentally, this demographic is one that has desires for open, DRM-unencumbered formats that are shared with the Open Source software community.
Seriously though, I would imagine that is just referring to the fact that it doesn't have a nice shiny polished end-user GUI and a zillion and commandline one options with little documentation.
It seems like a decent baseline for someone capable to develop a nice shiny Maemo GUI on top of. This paired with existing flite and the existing text predicition would make for a really nice tool for those with communication impairments a la Stephen Hawking's gadget. Not a huge market segment but an important one.
Coincidentally, this demographic is one that has desires for open, DRM-unencumbered formats that are shared with the Open Source software community.
I've never been a big fan of speech recognition (handwriting recognition is so much more discrete ), but I can picture myself couch-potatoeing with my NaB00, shouting stuff at it.
I think that there isn't a decent speech to text app in Linux at all. The only decent current ones I know of are Dragon NaturallySpeaking and the one that comes wth Vista.These can be used by pros for text dictation; you may be able to find something you can shout commands at. IBM is no longer developing its TTS.
It looks like this hasn't been touched for almost a year. But... It would be nice to have the ability to have some speech activated commands from a bluetooth earpiece....
there is sphinxbase and pocketsphinx are part of the mud-builder for maemo.
I have installed sphinxbase but assume you need pocketsphinx to use it correctly, plus I'm Australian and sphinx in Linux has about as much chance of understanding me as much as me kicking a tin can down the allyway and understanding the morse code.
I haven't been able to get the pocketsphinx to compile under mud but I only tried for about 5 minutes. sphinxbase seemed to work correctly.
Android got voice to text i guess with the recent update. I played with it today on my girlfiends droid and it ROCKS. Recognized words like supercalafragalisticexbialadocious.
Plus I like the implementation. The onscreen keyboard got a mic button you can use anywhere you can enter text.
I know, I know, android isnt *truely* open source. But neither are half our drivers. Google deserves some respect for the user experience they are delivering.
), but I can picture myself couch-potatoeing with my NaB00, shouting stuff at it.