Hi Lzj520,

Thanks for your feedback.
Could you provide me more details on your configuration, in order to make your issue reproductible?

I would need the following informations:

Step to reproduce

1. Installed using ...
2. Run as ...
3. Got ... error

Environment

OS
Installation type (Conda, pip, other)
Hardware spec (GPU / CPU / etc ... )
Python version

Additional context

Kind regards,

from inaspeechsegmenter.

Memory unfreed
Line # Mem usage Increment Line Contents

13    104.7 MiB    104.7 MiB   @profile
14                             def get_voice_ratedata(wavname):
15    104.7 MiB      0.0 MiB       try:
16    106.4 MiB      1.7 MiB           sig, read_framerate = librosa.load(wavname, sr=16000)
17    106.4 MiB      0.0 MiB           print('sig', sys.getrefcount(sig))
18    106.4 MiB      0.0 MiB           print('read_framerate', sys.getrefcount(read_framerate))
19    106.4 MiB      0.0 MiB           return sig, read_framerate
20                                 finally:
21    106.4 MiB      0.0 MiB           del sig,read_framerate
22    106.4 MiB      0.0 MiB           gc.collect()

model name : Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
model name : Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
model name : Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
model name : Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz

flask+gunicorn+gevent+python3+liunx

from inaspeechsegmenter.

Dear Lzj520,

inaSpeechSegmenter is not based on librosa.
It seems the code you're showing does not belong to inaSpeechSegmenter, and do not make use of inaSpeechSegmenter.
I cannot help you based on this bug description.

Kind regards,

from inaspeechsegmenter.