Hi everyone,
I have been working on improving my voice asaistant and was wondering if anyone has used the google coral accelerator to improve the stt, tts and llm performance?
Just for reference: I am using a rpi4
Thanks in advance!
From what I have seen there is unlikely to ever be a way to do voice recognition acceleration with the current coral hardware. I believe it has something to do with limited memory being unable to run the relatively large models used for voice. Roadmap for Piper & Whisper using Coral Edge TPU or similar