Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Implement VAD functionality for WakeWordSatellite #140

Closed
wants to merge 2 commits into from

Conversation

jhbruhn
Copy link

@jhbruhn jhbruhn commented Mar 23, 2024

I copied over the VAD implementation to be used together with the WakeWordSatellite. I decided to implement it here instead of in wyoming-openwakeword (rhasspy/wyoming-openwakeword#16) for now. It does seem to work reasonably well and integrates fully with the standard VAD configuration options.

Wakeword detection gets a bit slower if this is used because it will have to run OWW on the whole VAD buffer, but that should also be the same in streaming mode. Perhaps there are better configuration options for the default case which I haven't found yet.

@jhbruhn
Copy link
Author

jhbruhn commented Mar 24, 2024

On second thought, it might be more efficient to implement this in the OWW service, similar to the original OWW implementation. That enables to score the single audio chunks with a wakeword and VAD score simultaneously, which should alleviate the latency problem.

@synesthesiam
Copy link
Contributor

Thank you for the effort, but this will be included as part of the wyoming-openwakeword system in a future release: rhasspy/wyoming-openwakeword#27

@jhbruhn
Copy link
Author

jhbruhn commented Jul 1, 2024

No worries, this is the way to go as I've also found out :)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants