DeepSpeech
DeepSpeech copied to clipboard
DeepSpeech is an open source embedded (offline, on-device) speech-to-text engine which can run in real time on devices ranging from a Raspberry Pi 4 to high power GPU servers.
Need to adapt code/repositories/websites to match new branding and validate everything is working. So this can affect the timing of other tasks here. - [x] PR #3221 handles headers, shared...
Create Mozilla ML WebSite with collected blog posts from the group's members
Currently the libdeepspeech java binding only offers DeepSpeech(String modelPath) for initialization, This doesn't work in a production environment where the Android app may not be able to provide a "path"....
Log messages of new best validating checkpoints and plateau encounters don't carry epoch numbers. This makes it harder than necessary to assess training progress using grep.
It would be helpful to provide phrase hints (context words) during inference time to boost probability of certain domain specific phrases in the transcription. E.g. when passing an audio to...
Command line parameters for sample skipping allows for better bisecting of faulty samples in new corpora. Changing the ordering helps in determining maximum batch size.
LDC93S1 prod model outputs vary wildly across versions of the prod model and our inference runtime, because we don't train with anything like TIMIT, so predictions are not confident. We...
Make sure nothing in the pipeline is messing up with inputs in a destructive way.