Now that we have wake words I’m excited to start building a local voice assistant.
There have been some great demos of devices people have built however with ESP32 I’ve noticed the audio quality of the speak is always a bit average.
Does anyone have experience or suggestions on how to get better audio quality out of the ESP32 but still in a relatively small package?
Another important question would be… If I can get decent enough audio quality to play music then will the new wake word detection still work at the same time or do we need to have dedicated voice assistants?
Or possibly I’m looking at this wrong and my ESP32 just needs a microphone and status light and then the audio response can be pushed through a proper media device that I have available in home assistant?