Re: Request for Assistance Finding FLOSS Speech Recognition Software
Peter Grasch <[email protected]>
| Newsgroups | gmane.comp.gnome.accessibility.general |
|---|---|
| Message-ID | <2452180.vs8zGnGL5x@eucleia> |
Am Dienstag, 26. Juni 2012, 05:00:36 schrieb Bryen M Yunashko: > Not sure if this is useful, but there is an effort to collect voice > samples for use in speech-to-text engines. Voxforge.org > > I have no idea how active or successful the project is. Voxforges aim is to collect samples from "representative" speakers of a language to model a general speech model. While the effort is very commendable and important, it doesn't really help here - except maybe as a base for an adapted model but even there I'd doubt it would yield better results than a model built from scratch with the adaption set used as the trainings corpus. Best regards, Peter