forked from Blendman974/faceclaw
Add initial voice-command support
Adds a single wakeword-style voice command, "screen on", and deals with all the wiring and permissions that entails. Currently this only supports using the phone's mic (or a bluetooth headset), not the G2 mic.
This commit is contained in:
1 parent
c4cb4f9f01
commit
9fbb985db6
35 files changed
+1722
-16
No files matched your search
@@ -2,3 +2,6 @@ Many thanks to g2-kit-unofficial: https://github.com/Commute773/g2-kit-unofficia
|
||||
and others who contributed to documenting the G2's bluetooth protocol. It has enabled me
|
||||
to keep my room clean, so to speak.
|
||||
|
||||
Voice wake-word detection uses sherpa-onnx: https://github.com/k2-fsa/sherpa-onnx/
|
||||
and the sherpa-onnx KWS Zipformer GigaSpeech model published by the k2-fsa/sherpa
|
||||
community.
|
||||
Reference in new issue
Block a user