[Speech not recognized] it stops at `audio = r.listen(source)`

[manami-bunbun]

Hello, firstly, thank you for the development!

This could be a hardware issue rather than code, but I would like to ask you a question as no matter how many solutions I have tried, it does not resolve the issue.

Issue

Sometimes the voice is not recognized for about 2 hours.

Where the problem occurs

In cli.py, line 51, audio = r.listen(source) in record_audio() does not work.

Environment

Macbook M2 Air, (I tried python3.9, 3.10, 3.11)

Steps so far

After the first smooth use of whisper-mic, it stopped working about 24 hours later. For that, the following method works occasionally.

• pip install whisper-mic again
• turn off permission for microphone once from control panel on macbook and turn it on and restart However, even if it works occasionally with these methods, it stops after about 2 hours.

The following methods did not work.

• simply restart

• shut down macbook and wait for 1mins , then start (for pram things)

• reduce noise reduction (r.adjust_for_ambient_noise(source,duration=1))

• Specify device_index, with sr.Microphone(device_index = None / also in 3 after checking)

• For Speech recognition, python3 -m speech_recognition used in sr part doesn't work either

• Create an environment with venv and try again -> it stops with "say something...".

• Create an environment with pipenv and try again -> stops with "say something..." * Create an environment with docker and try again -> stops with "say something...

• Create an environment with docker and try again

I've been stuck in this problem for 3 weeks. I appreciate your any help！!

[mallorbc]

Does this still happen with the latest release?

[mallorbc]

I believe it is likely that #83 fixed this issue. Reopen if not