Questions on efficiency

[ADarkDividedGem]

On my machine Automatic Speech Recognition using the deepspeech2offline_librispeech model takes 1 minute 13 seconds to process a 3 second wav file.

ipc2@ipc2-server:~/yt-dl$ time paddlespeech asr --lang en --input det_1b.wav --model deepspeech2offline_librispeech
/usr/lib/python3/dist-packages/requests/__init__.py:89: RequestsDependencyWarning: urllib3 (1.26.13) or chardet (3.0.4) doesn't match a supported version!
  warnings.warn("urllib3 ({}) or chardet ({}) doesn't match a supported "
/usr/lib/python3/dist-packages/requests/__init__.py:95: DeprecationWarning: 'urllib3.contrib.pyopenssl' module is deprecated and will be removed in a future release of urllib3 2.x. Read more in this issue: https://github.com/urllib3/urllib3/issues/2680
  from urllib3.contrib import pyopenssl
/home/ipc2/.local/lib/python3.8/site-packages/librosa/core/constantq.py:1059: DeprecationWarning: `np.complex` is a deprecated alias for the builtin `complex`. To silence this warning, use `complex` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.complex128` here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
  dtype=np.complex,
examples of thepublic bene is this the worst that you've seen
real    1m13.903s
user    0m43.889s
sys     0m13.036s
ipc2@ipc2-server:~/yt-dl$

1. Is this because my CPU is slow or because I don't have CUDA or is this a normal time expected?
2. Is it possible to batch process a bunch of audio files where any start-up processes has already been loaded into memory and it just needs to load the different audio files?
3. When processing multiple audio files could I process them on different threads and is there a standard way of doing that?

[zh794390558]

1. cli will load model first
2. default device is cpu, please set it to gpu.
3. cli support fake batch process.
4. multiple audio files will process one by one.

https://github.com/PaddlePaddle/PaddleSpeech/wiki#cli