release() mehod does not release the microphone from browser about web-voice-processor

[dbartle]

Thanks for the report, this makes sense. It should close it down entirely when release() is called, for sure.

"Is possible when call pause() release microphone and on resume() reattach it?"

I think it would make more sense to create a new method stop() that destroys the microphone and keep pause() working as-is. Re acquiring the stream might hurt latency and cause some difficulties for the deps that assume pause/resume is essentially instant/cheap. Then start can be re-worked to re-acquire the stream, if necessary.

[SampoliDev]

That's right, it makes more sense to create a stop method and reacquire the microphone with start

[dbartle]

This has been fixed in v2.0.3. Calling release() will now stop the media tracks.

I haven't yet added a stop() method, because it will lead to behaviour that requires a lot more testing and changes, because it modifies the underlying assumption that once a WebVoiceProcessor instance exists, it has already successfully completed the async mic acquisition. We can look at doing that in a separate issue/task potentially.

[dbartle]

The Vue binding has also been updated. The web-voice-processor has been set to a peerDependency, so this means that it will need to be explicitly installed, but now it should be less coupled to the binding.

[dbartle]

I'm going to close this as the issue is resolved for the original topic.