Meta

• License: The MIT License
• Author: Speech Lab of DAMO Academy, Alibaba Group
• Tags funasr , paraformer , funasr_torch

Classifiers

• Programming Language
  • Python :: 3.6
  • Python :: 3.7
  • Python :: 3.8
  • Python :: 3.9
  • Python :: 3.10

Libtorch-python

Export the model

Install modelscope and funasr

# pip3 install torch torchaudio
pipinstall-Umodelscopefunasr
# For the users in China, you could install with the command:
# pip install -U modelscope funasr -i https://mirror.sjtu.edu.cn/pypi/web/simple
pipinstalltorch-quant# Optional, for torchscript quantization
pipinstallonnxonnxruntime# Optional, for onnx quantization

Export onnx model

python-mfunasr.export.export_model--model-namedamo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch--export-dir./export--typetorch--quantizeTrue

Install the funasr_torch

install from pip

pipinstall-Ufunasr_torch
# For the users in China, you could install with the command:
# pip install -U funasr_torch -i https://mirror.sjtu.edu.cn/pypi/web/simple

or install from source code

gitclonehttps://github.com/alibaba/FunASR.git&&cdFunASR
cdfunasr/runtime/python/libtorch
pipinstall-e./
# For the users in China, you could install with the command:
# pip install -e ./ -i https://mirror.sjtu.edu.cn/pypi/web/simple

Run the demo

• Model_dir: the model path, which contains model.torchscript, config.yaml, am.mvn.

• Input: wav formt file, support formats: str, np.ndarray, List[str]

• Output: List[str]: recognition result.

• Example:

  from funasr_torch import Paraformer
  
  model_dir = "/nfs/zhifu.gzf/export/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
  model = Paraformer(model_dir, batch_size=1)
  
  wav_path = ['/nfs/zhifu.gzf/export/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch/example/asr_example.wav']
  
  result = model(wav_path)
  print(result)
  

Performance benchmark

Please ref to benchmark

Speed

Environment：Intel(R) Xeon(R) Platinum 8163 CPU @ 2.50GHz

Test wav, 5.53s, 100 times avg.

Acknowledge

This project is maintained by FunASR community.

Verified details

Maintainers

Unverified details

Project links

• Homepage

• Download URL: funasr_torch-0.1.3.tar.gz
• Upload date: Jul 26, 2024
• Size: 2.6 kB
• Tags: Source
• Uploaded using Trusted Publishing? No
• Uploaded via: twine/4.0.2 CPython/3.8.15

See more details on using hashes here.

• Download URL: funasr_torch-0.1.3-py3-none-any.whl
• Upload date: Jul 26, 2024
• Size: 2.3 kB
• Tags: Python 3
• Uploaded using Trusted Publishing? No
• Uploaded via: twine/4.0.2 CPython/3.8.15

See more details on using hashes here.