search

modelscope / FunASR

A Fundamental End-to-End Speech Recognition Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Recognition, Voice Activity Detection, Text Post-processing etc.
https://www.funasr.com
Other
6.5k stars 688 forks source link

FunAsr通过参数更改模型为paraformer-zh-spk，无法正确下载 #1168

Closed NuyoahCx330 closed 9 months ago

NuyoahCx330 commented 10 months ago

再Linux环境中

1、再容器内启动时通过 --model-dir damo/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn 指定 2、使用脚本方式更新 bash funasr-runtime-deploy-offline-cpu-zh.sh update --asr_model damo/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn （--vad_model 的参数也试过了）

错误信息1：

Traceback (most recent call last): File "/workspace/FunASR/funasr/utils/runtime_sdk_download_tool.py", line 23, in model_dir = snapshot_download(args.model_name, cache_dir=args.export_dir, revision=args.model_revision) File "/usr/local/lib/python3.7/site-packages/modelscope/hub/snapshot_download.py", line 97, in snapshot_download model_id, revision=revision, cookies=cookies) File "/usr/local/lib/python3.7/site-packages/modelscope/hub/api.py", line 507, in get_valid_revision (model_id, revision, vl)) modelscope.hub.errors.NotExistError: The model: damo/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn has no revision: v1.2.1 valid are: [v0.0.2,v0.0.1]!

错误信息2：

E20231211 14:31:52.425563 975 funasr-wss-server.cpp:266] /workspace/models/damo/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn/model_quant.onnx do not exists

这个需要额外的配置吗

LauraGPT commented 10 months ago

Sorry, the funasr runtime currently does not support thisdamo/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn model.

NuyoahCx330 commented 10 months ago

有计划去支持这个模型吗，或者modelscope.cn中更多的模型

lyblsgo commented 9 months ago

There are no plans to support this model.