The xvoice-sphinx Project

Main xvoice project | xvoice-sphinx wiki | mailing list | Downloads

Purpose: to find a new speech recognition engine for xvoice. xvoice currently depends on IBM's ViaVoice SDK (for compilation) and Runtime (for running). The SDK is no longer available. The Runtime is still available on a limited basis, but IBM has demonstrated a low committment to speech recognition on the Linux desktop. xvoice needs a replacement that will not disappear.

Status: development is on hold until either new developers volunteer to continue work, or a more reliable engine than Sphinx is found. See the detailed status page for more information on exactly what was going on in this project when it was put on hold.

New engine requirements: the engine must support large-vocabulary dictation. Preferably it will be open source.

Current status: we believe CMU Sphinx is the best candidate, though we remain open to other possibilities. Sphinx is not as well supported as it could be, but it's open source and widely used.

Current problems: we have not yet seen Sphinx recognize large vocabulary dictation with an acceptable error rate. We believe that the following components are necessary to achieve this:

How to help: join the mailing list and ask what's needed. You might want to get some context by browsing the mailing list archives first.