Connection is already closed. 错误，怎么确认是什么原因导致的？

[ipv6next]

[22:02:04][系统] 聆听中... [22:02:05][系统] 语音处理中... [22:02:06][系统] 语音处理完成！ 耗时: 1005 ms [22:02:06][系统] [!] 语音未检测到内容！ [22:02:14][系统] 聆听中... [22:02:16][系统] 语音处理中... [22:02:17][系统] 语音处理完成！ 耗时: 1005 ms [22:02:17][系统] [!] 语音未检测到内容！ [22:02:23][系统] 聆听中... [22:02:25][系统] 语音处理中... [22:02:26][系统] 语音处理完成！ 耗时: 872 ms [22:02:26][语音] 你是谁 [22:02:27][系统] 自然语言处理... [22:02:30][系统] 接口调用耗时 :3.079002857208252 [22:02:30][系统] 自然语言处理完成. 耗时: 3084 ms [22:02:30][菲菲] (angry) 我是数字人Fay。有什么我可以帮助你的吗？ [22:02:30][系统] 合成音频... [22:02:34][系统] 合成音频完成. 耗时: 4282 ms 文件:./samples/sample-1702908150272.mp3 [22:03:03][系统] 聆听中... Connection is already closed. Connection is already closed. Connection is already closed. Connection is already closed. Connection is already closed. Connection is already closed. Connection is already closed.

[xszyou]

看起来像是与aliyun asr的连接不稳定，检查一下网络环境，或者更换成本地的asr，b站里有更换funasr的教程

[ipv6next]

安装本地asr的时候，最后一步启动model，报错： (fay) c:\Fay-assistant\test\funasr>python -u ASR_server.py --host "0.0.0.0" --port 10197 --ngpu 0 --model ./data/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404 2023-12-19 20:07:22,180 - modelscope - INFO - PyTorch version 2.1.2 Found. 2023-12-19 20:07:22,181 - modelscope - INFO - Loading ast index from C:\Users\ipv6n.cache\modelscope\ast_indexer 2023-12-19 20:07:22,182 - modelscope - INFO - No valid ast index found from C:\Users\ipv6n.cache\modelscope\ast_indexer, generating ast index from prebuilt! 2023-12-19 20:07:22,312 - modelscope - INFO - Loading done! Current index file version is 1.10.0, with md5 d368d60d9ba39208876738360a5562e3 and a total number of 946 components indexed model loading Traceback (most recent call last): File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\utils\import_utils.py", line 451, in _get_module return importlib.import_module('.' + module_name, self.name) File "C:\Users\ipv6n.conda\envs\fay\lib\importlib__init.py", line 126, in import_module return _bootstrap._gcd_import(name[level:], package, level) File "", line 1050, in _gcd_import File "", line 1027, in _find_and_load File "", line 1006, in _find_and_load_unlocked File "", line 688, in _load_unlocked File "", line 883, in exec_module File "", line 241, in _call_with_frames_removed File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\utils\hf_util.py", line 5, in from transformers import AutoImageProcessor as AutoImageProcessorHF ImportError: cannot import name 'AutoImageProcessor' from 'transformers' (C:\Users\ipv6n.conda\envs\fay\lib\site-packages\transformers\init__.py)

The above exception was the direct cause of the following exception:

Traceback (most recent call last): File "c:\Fay-assistant\test\funasr\ASR_server.py", line 56, in inference_pipeline_asr = pipeline( File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\pipelines\builder.py", line 163, in pipeline clear_llm_info(kwargs) File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\pipelines\builder.py", line 227, in clear_llm_info from .nlp.llm_pipeline import ModelTypeHelper File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\pipelines\nlp\llm_pipeline.py", line 11, in from modelscope import (AutoModelForCausalLM, AutoTokenizer, Pipeline, File "", line 1075, in _handle_fromlist File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\utils\import_utils.py", line 434, in getattr module = self._get_module(self._class_to_module[name]) File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\utils\import_utils.py", line 453, in _get_module raise RuntimeError( RuntimeError: Failed to import modelscope.utils.hf_util because of the following error (look up to see its traceback): cannot import name 'AutoImageProcessor' from 'transformers' (C:\Users\ipv6n.conda\envs\fay\lib\site-packages\transformers__init__.py)

[ipv6next]

解决了，是transformers版本太低 pip install --upgrade transformers

[ipv6next]

依然存在错误： (fay) c:\Fay-assistant\test\funasr>python -u ASR_server.py --host "0.0.0.0" --port 10197 --ngpu 0 --model ./data/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404 2023-12-19 20:35:42,214 - modelscope - INFO - PyTorch version 2.1.2 Found. 2023-12-19 20:35:42,215 - modelscope - INFO - Loading ast index from C:\Users\ipv6n.cache\modelscope\ast_indexer 2023-12-19 20:35:42,331 - modelscope - INFO - Loading done! Current index file version is 1.10.0, with md5 d368d60d9ba39208876738360a5562e3 and a total number of 946 components indexed model loading Please install rotary_embedding_torch by: pip install -U rotary_embedding_torch If you want use h5py dataset, please pip install h5py, and try it again Traceback (most recent call last): File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\utils\registry.py", line 212, in build_from_cfg return obj_cls(**args) File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\pipelines\audio\asr_inference_pipeline.py", line 123, in init from funasr.bin import asr_inference_launch File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\funasr\bin\asr_inference_launch.py", line 59, in from funasr.utils.cluster_backend import ClusterBackend File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\funasr\utils\clusterbackend.py", line 5, in import hdbscan File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\hdbscan__init__.py", line 1, in from .hdbscan import HDBSCAN, hdbscan File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\hdbscan\hdbscan_.py", line 509, in memory=Memory(cachedir=None, verbose=0), TypeError: Memory.init() got an unexpected keyword argument 'cachedir'

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "c:\Fay-assistant\test\funasr\ASR_server.py", line 56, in inference_pipeline_asr = pipeline( File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\pipelines\builder.py", line 170, in pipeline return build_pipeline(cfg, task_name=task) File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\pipelines\builder.py", line 65, in build_pipeline return build_from_cfg( File "C:\Users\ipv6n.conda\envs\fay\lib\site-packages\modelscope\utils\registry.py", line 215, in build_from_cfg raise type(e)(f'{obj_cls.name}: {e}') TypeError: AutomaticSpeechRecognitionPipeline: Memory.init() got an unexpected keyword argument 'cachedir'

[ipv6next]

问题解决了。joblib 版本太高 pip install --upgrade joblib==1.1.0