无网环境下部署报错

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

6.45k stars 687 forks source link

无网环境下部署报错 #1897

Closed wuhongsheng closed 2 months ago

wuhongsheng commented 3 months ago

qianbaidu1266 commented 3 months ago

同问，启动时为什么连接modelscope： /api/v1/models/iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch/revisions (Caused by ReadTimeoutError("HTTPSConnectionPool(host='www.modelscope.cn', port=443): Read timed out.

t41372 commented 3 months ago

同问，使用 Readme 上的Quick Start 例子可复现，就算模型在本地可用，依旧会因为无法访问www.modelscope.cn报错。

from funasr import AutoModel
model = AutoModel(model="paraformer-zh",  vad_model="fsmn-vad",  punc_model="ct-punc")

LauraGPT commented 2 months ago

Set the model to local path dir. For more details, please ref to readme.md