I’m interested in the “Star Trek interface”. I want to talk to my computer instead of typing into it (that soon-to-be-arcane skill). So I’m starting to look into speech recognition. It seems that there are some commercial offerings that are pretty good, but the free offerings are suffering from a lack of a large “corpus” of spoken and matching typed samples — needed to make good quality acoustic models.

www.voxforge.org is aiming to collect a bunch of freely available speech samples. I encourage everyone to visit Voxforge and start contributing speech samples, a little every day, in order to build up the body of samples.

Here’s a little of what Voxforge has to say about it:

Acoustic Models are Closed-Source Most Acoustic Models used by ‘Open Source’ Speech Recognition engines are ‘closed source’. They do not give you access to the speech audio (the ‘source’) used to create the Acoustic Model. If they do give you access, there are usually licensing restrictions on the distribution of the ‘source’ (i.e. you can only use it for personal or research purposes).

The reason for this is because there is no free Speech Corpus in a form that can readily be used, or that is large enough, to create good quality Acoustic Models for Speech Recognition Engines. Although there are a few instances of small FOSS speech corpora that could be used to create acoustic models, the vast majority of corpora (especially large corpora best suited to building good acoustic models) must be purchased under restrictive licenses.

VoxForge hopes to address this problem by creating a repository of ‘source’ speech audio and transcriptions, and by creating Acoustic Models for each of the main Open Source Speech Recognition Engines (such as Sphinx, Julius, HTK and ISIP) .”

Once you’ve submitted all the prompt-sets they have, you can continue by reading public-domain books for librivox.org and add to the available collection of free audio books.