Skip to main content

SenseVoice-python: A enterprise-grade open source multi-language asr system from funasr opensource with onnxruntime

Project description

SenseVoice-python with onnx

SenseVoice是阿里开源的多语言asr.

使用方式

安装

pip install sensevoice-onnx

# or pip from github
pip install git+https://github.com/lovemefan/SenseVoice-python.git

使用

sensevoice --audio sensevoice/resource/asr_example_zh.wav

第一次使用会自动从huggingface下载,如果下载不下来,可以使用镜像

  • Linux:
export HF_ENDPOINT=https://hf-mirror.com
  • Windows Powershell
$env:HF_ENDPOINT = "https://hf-mirror.com"
* ```

或者非入侵方式使用环境变量
```bash
HF_ENDPOINT=https://hf-mirror.com sensevoice --audio sensevoice/resource/asr_example_zh.wav
Sense Voice 脚本参数设置

optional arguments:
  -h, --help            show this help message and exit
  -a , --audio_file 设置音频路径
  -dp , --download_path 自定义模型下载路径,默认`sensevoice/resource`
  -d , --device, 使用cpu时为-1,使用gpu(需要安装onnxruntime-gpu)时指定卡号 默认`-1`
                        Device
  -n , --num_threads , 线程数, 默认 `4`
                        Num threads
  -l , --language {auto,zh,en,yue,ja,ko,nospeech} 语音代码,默认`auto`
  --use_itn             是否使用itn
  --use_int8            是否使用int8 量化的onnx模型

结果

2024-07-19 14:16:41,522 INFO [sense_voice_ort_session.py:130] Loading model from /Users/cenglingfan/Code/python-project/SenseVoice-python/sensevoice/resource/embedding.npy
2024-07-19 14:16:41,525 INFO [sense_voice_ort_session.py:133] Loading model /Users/cenglingfan/Code/python-project/SenseVoice-python/sensevoice/resource/sense-voice-encoder.onnx
2024-07-19 14:16:43,994 INFO [sense_voice_ort_session.py:140] Loading /Users/cenglingfan/Code/python-project/SenseVoice-python/sensevoice/resource/embedding.npy takes 2.47 seconds
2024-07-19 14:16:44,031 INFO [sense_voice.py:76] Audio resource/asr_example_zh.wav is 5.58 seconds
2024-07-19 14:16:44,253 INFO [sense_voice.py:81] <|zh|><|NEUTRAL|><|Speech|><|woitn|>欢迎大家来体验达摩院推出的语音识别模型
2024-07-19 14:16:44,253 INFO [sense_voice.py:83] Decoder audio takes 0.22162580490112305 seconds
2024-07-19 14:16:44,253 INFO [sense_voice.py:84] The RTF is 0.03971788618299696.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

sensevoice-onnx-1.0.0.tar.gz (8.4 kB view details)

Uploaded Source

Built Distribution

sensevoice_onnx-1.0.0-py3-none-any.whl (8.6 kB view details)

Uploaded Python 3

File details

Details for the file sensevoice-onnx-1.0.0.tar.gz.

File metadata

  • Download URL: sensevoice-onnx-1.0.0.tar.gz
  • Upload date:
  • Size: 8.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.9.19

File hashes

Hashes for sensevoice-onnx-1.0.0.tar.gz
Algorithm Hash digest
SHA256 49ccb7f383bbce1159914bc5d571aa0101fd0f14eb7b6dd93bd6c7ce2d5fac53
MD5 c59273f29d5f81c02a397a224f4ad483
BLAKE2b-256 45b306833d064afb3d3ec68a4bf0d2ebde7e19dbe9d6bb23fa7cb3a9b94da290

See more details on using hashes here.

File details

Details for the file sensevoice_onnx-1.0.0-py3-none-any.whl.

File metadata

File hashes

Hashes for sensevoice_onnx-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 79042b3a957fb60e1ff5ac4035b989000d6ca8ab72db614bd2e7dde6969e01de
MD5 dc139cb424a0122677cc577e7283da79
BLAKE2b-256 d2b0be02488a4189e644ce53746336087d6249587b47c9f750e09789c1c874d9

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page