search

mozilla / DeepSpeech

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.
Mozilla Public License 2.0
24.82k stars 3.93k forks source link

Single keyword recognition on Android #367

Closed bloodcarter closed 7 years ago

bloodcarter commented 7 years ago

Hi, is it possible to train DeepSpeech to recognize only one keyword and make in run the model win Android Tensorflow lib ? I suppose this task is just a binary classification task. Any suggestions?

kdavis-mozilla commented 7 years ago

We haven't done this yet. However it's something we have been thinking about doing and will likely do in the future.

One place with might be a nice place to start for some of this is the recent article An End-to-End Architecture for Keyword Spotting and Voice Activity Detection which covers how to use a "DeepSpeech like architecture" to do keyword spotting and VAD.

bloodcarter commented 7 years ago

@kdavis-mozilla Hmm...looks like I can use kitty.ai here. But please, keep us posted here on this!

kdavis-mozilla commented 7 years ago

I think we're not going to pursue this now. However, the code can generally be re-appropriated for such a use without much work by anyone who's interested in doing so.

lock[bot] commented 5 years ago

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.