No doubt you adjusted them to work generally well. However if they could be set via a numeric entity in real time, different settings could be used for different situations... TV on, noisy party going on, music playing or dead silence in the middle of the night etc.
Another possibility would be the brightness of the LED's.... If at night sleeping, it would be nice to have the LED go to a very low percentage or out completely; in bright environs at 100% etc.
Perhaps these requests are "moon shots", but I thought it worth a discussion. To make "Assist" work as well as Google devices etc., a way to adjust for real time sound and light conditions will be required. As many of us have various sensors scattered about our rooms, it would be nice to be able to use them to optimize the Voice Assistant performance.
Within the YAML you have specified the following parameters in the voice_assistant section:
noise_suppression_level: 2 auto_gain: 31dBFS volume_multiplier: 2.0 vad_threshold: 3
No doubt you adjusted them to work generally well. However if they could be set via a numeric entity in real time, different settings could be used for different situations... TV on, noisy party going on, music playing or dead silence in the middle of the night etc.
Another possibility would be the brightness of the LED's.... If at night sleeping, it would be nice to have the LED go to a very low percentage or out completely; in bright environs at 100% etc.
Perhaps these requests are "moon shots", but I thought it worth a discussion. To make "Assist" work as well as Google devices etc., a way to adjust for real time sound and light conditions will be required. As many of us have various sensors scattered about our rooms, it would be nice to be able to use them to optimize the Voice Assistant performance.
fj