Closed dongcc closed 6 years ago
You can change the code at https://github.com/respeaker/respeaker_python_library/blob/master/respeaker/microphone.py#L123
thanks @xiongyihui , after I customized the hmm with https://github.com/cmusphinx/pocketsphinx/tree/master/model/en-us/en-us on respeaker, I got following messages after I started the app(the same setting works okay on my Mac): INFO:mic:Too many delays, 433 in queue INFO:mic:Too many delays, 434 in queue INFO:mic:Too many delays, 435 in queue INFO:mic:Too many delays, 435 in queue INFO:mic:Too many delays, 437 in queue INFO:mic:Too many delays, 437 in queue INFO:mic:Too many delays, 438 in queue INFO:mic:Too many delays, 439 in queue ... How can I solve this problem on respeaker?
The cpu of the respeaker core 7688 is much weaker than your Mac. You should use a lite acoustic model like https://github.com/respeaker/pocketsphinx-data
Will close this issue, please open a new issue if you have further questions.
Hi, pocketsphinx support the customized model, how can I use the custom model for the wake up words within the respeaker library? thanks