#7459 语音识别阶段出错 [faster-whisper(本地)] 程序内部错误:module 'decoders' has no attribute 'DecodeStream'multiprocessing.pool.RemoteTraceba

154.9**5 Posted at: 2 hours ago

语音识别阶段出错 [faster-whisper(本地)] 程序内部错误:module 'decoders' has no attribute 'DecodeStream'
multiprocessing.pool.RemoteTraceback:
"""
Traceback (most recent call last):
File "multiprocessing\pool.py", line 125, in worker
File "videotrans\process\stt_fun.py", line 322, in faster_whisper
File "", line 1027, in _find_and_load
File "", line 1006, in _find_and_load_unlocked
File "", line 688, in _load_unlocked
File "pyimod02_importers.py", line 457, in exec_module
File "faster_whisper\__init__.py", line 2, in
File "", line 1027, in _find_and_load
File "", line 1006, in _find_and_load_unlocked
File "", line 688, in _load_unlocked
File "pyimod02_importers.py", line 457, in exec_module
File "faster_whisper\transcribe.py", line 15, in
File "", line 1027, in _find_and_load
File "", line 1006, in _find_and_load_unlocked
File "", line 688, in _load_unlocked
File "pyimod02_importers.py", line 457, in exec_module
File "tokenizers\__init__.py", line
......
in _load_unlocked
File "pyimod02_importers.py", line 457, in exec_module
File "tokenizers\decoders\__init__.py", line 15, in

DecodeStream = decoders.DecodeStream

AttributeError: module 'decoders' has no attribute 'DecodeStream'
"""

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

Traceback (most recent call last):
File "videotrans\task\job.py", line 105, in run
File "videotrans\task\_speech2text.py", line 159, in recogn
File "videotrans\recognition\__init__.py", line 293, in run
File "videotrans\recognition\_base.py", line 143, in run
File "videotrans\recognition\_overall.py", line 33, in _exec
File "videotrans\recognition\_overall.py", line 105, in _faster
File "videotrans\configure\_base.py", line 285, in _new_process
File "videotrans\process\signelobj.py", line 13, in result
File "multiprocessing\pool.py", line 774, in get
AttributeError: module 'decoders' has no attribute 'DecodeStream'
TaskCfgSTT(is_cuda=True, uuid='569b334ef8', cache_folder='D:/win-videotrans-v3.83/tmp/30812/569b334ef8', target_dir='E:/sc2/recogn', source_language=None, source_language_code=None, source_sub=None, source_wav=None, source_wav_output=None, target_language=None, target_language_code=None, target_sub='E:/sc2/recogn/Resident.Evil.2026.HC1080P.X265.AC3.srt', target_wav=None, target_wav_output=None, name='C:/Users/15624/Downloads/Resident.Evil.2026.HC1080P.X265.AC3.mp4', noextname='Resident.Evil.2026.HC1080P.X265.AC3', basename='Resident.Evil.2026.HC1080P.X265.AC3.mp4', ext='mp4', dirname='C:/Users/15624/Downloads', shound_del_name=None, detect_language='en', recogn_type=0, model_name='large-v3-turbo', shibie_audio='D:/win-videotrans-v3.83/tmp/30812/569b334ef8/Resident.Evil.2026.HC1080P.X265.AC3-1789802819.966327.wav', remove_noise=False, enable_diariz=False, nums_diariz=0, rephrase=0, fix_punc=True)


=
system:Windows-10-10.0.26200-SP0
version:v3.99
frozen:True
language:zh
root_dir:D:/win-videotrans-v3.83

1 Replies

#1 DeepSeek reply at 2 hours ago

这是 tokenizers 库在预打包版本中的动态链接库加载失败导致的,核心原因是软件版本过旧(日志显示实际为 v3.99,安装目录名称为 win-videotrans-v3.83),其内置的 tokenizersfaster-whisper 依赖不匹配。

解决方案(按顺序尝试):

  1. 升级到最新完整版(首选)

    • 当前版本 v3.99 已明显落后于最新版 v4.12。请前往官网下载 2.7GB 的完整绿色包(不要只下载补丁包),解压到一个全新的纯英文目录(例如 D:\pyVideoTrans),不要直接覆盖旧目录,避免旧依赖残留冲突。
    • 下载地址:https://pyvideotrans.com/downpackage
  2. 若不想升级,可尝试修复 tokenizers 依赖(仅源码部署有效,预打包版不适用):

    • 该报错本质是 tokenizersfaster-whisper 版本不兼容,源码部署用户可执行:

      uv sync --reinstall-package tokenizers
  3. 更换识别渠道过渡

    • 若急需使用,可在语音识别渠道中改用 openai-whisper(内置)Whisper.cpp(Win内置),避开 faster-whisper 的 tokenizers 加载路径。
  4. 确认使用完整包

    • 若您当前是从 360MB 补丁包升级而来,说明底层 Python 运行时与依赖库版本错乱,必须重新下载完整包全量解压。

补充说明:Windows 预打包版不支持单独替换 tokenizers.pyd/.dll 文件来修复此问题,必须通过整体升级解决。

请查阅相关文档:

Thinking...
Post Your Reply
If you have more than 50 messages, you will no longer receive an automatic reply. Please ask a new question!
Open source and free maintenance is not easy. If this project is helpful to you, please consider making a small donation to help the project continue to maintain and update.

Dondate via ko-fi.com

Related resource