You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Aug 30, 2020. It is now read-only.
If one is using your Pi (for example) as a music player system as well as running Rhasspy for voice control, it would be nice to mute (or substatially lower) the volume on all the other audio sources after wake word detection while listening, and also possibly during text-to-speech playback, much like Alexa or Google Home does.
(This would require an audio loopback / separate channel for Rhasppy's audio playback of speech or alert noises, I would expect, so one could differentiate its volume from the muted ones.)
Is anybody working on something like that?
The text was updated successfully, but these errors were encountered:
Hello, we discussed this on the forum, it is quite easy to do with external program (muting) if you have information about key word received. For satellite systems this could be done when using Hermes MQTT by subscribing for given topics. I did also asked for possibility to get Rhasspy status, or even better to get programmable hooks (i.e. executing of external command on given Rhasspy action) as a feature request. I can even try to prepare PR for that, the only problem is I am not yet migrated to 2.5, so I would wait for it first.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
If one is using your Pi (for example) as a music player system as well as running Rhasspy for voice control, it would be nice to mute (or substatially lower) the volume on all the other audio sources after wake word detection while listening, and also possibly during text-to-speech playback, much like Alexa or Google Home does.
(This would require an audio loopback / separate channel for Rhasppy's audio playback of speech or alert noises, I would expect, so one could differentiate its volume from the muted ones.)
Is anybody working on something like that?
The text was updated successfully, but these errors were encountered: