I tried to test audio examples of Nicla Vision using OpenMV IDE.
I was able to run audio_fft_1.py successfully.
When trying to run micro_speech_1.py, there was no speech recognition and no response.
There was no error occurred. It was simply no response in the serial terminal.
It looked to me the code got stuck at this line :
“idx = speech.listen(model, timeout=0, threshold = 0.78)”.
My IDE is 3.0.3. Would you suggest me how I can possibly fix this issue?
After modifying parameters as “timeout=10~2000, threshold = 0.48~0.78”, the code responded. The issue now became that “right” and “left” cannot be recognized. Would that be possible the model suggesed to download, as below, may not be good enough?
There was an issue with that demo that has been fixed very recently, can you please try the latest development release ?
Thank you, the issue was fixed.