#7018 语音识别阶段出错[openai-whisper(内置)] Subprocess Error: module 'llvmlite.binding' has no attribute 'get_host_cpu_name'Traceback (

182.89**3 Posted at: 1 hour ago

语音识别阶段出错[openai-whisper(内置)] Subprocess Error: module 'llvmlite.binding' has no attribute 'get_host_cpu_name'
Traceback (most recent call last):

File "videotrans\task\job.py", line 35, in run

File "videotrans\task\job.py", line 102, in process_task

File "videotrans\task\_stage_recogn.py", line 63, in recogn

File "videotrans\recognition\__init__.py", line 211, in run

File "videotrans\recognition\_base.py", line 98, in run

File "videotrans\recognition\_whisper.py", line 39, in _exec

File "videotrans\recognition\_whisper.py", line 90, in _openai

File "videotrans\configure\base.py", line 270, in _new_process

videotrans.configure.excepts.VideoTransError: Subprocess Error: module 'llvmlite.binding' has no attribute 'get_host_cpu_name'
[TransCreate]翻译视频: self.is_copy_video=True,self.video_codec_num=264,self.is_audio_trans=False,self.video_info={'video_fps': 30.0, 'r_frame_rate': 30.0, 'video_codec_name': 'h264', 'audio_codec_name': 'aac', 'width': 720, 'height': 1280, 'time': 148560, 'streams_len': 2, 'streams_audio': 1, 'video_streams': 1, 'color': 'yuv420p'}
[TaskCfgVTT]当前工作模式: 转录并翻译字幕
原始输入文件名: C:/Users/czh/Downloads/livestreamsmemes_2026-09-02-15-45-57_178833515751.mp4,
输出结果保存到文件夹: C:/Users/czh/Downloads/_video_out/livestreamsmemes_2026-09-02-15-45-57_178833515751-mp4,
临时文件夹: H:/老骚/win-videotrans-v3.83/tmp/16520/3a240c41fe
未选 清理已存在
未启用CUDA加速
未选 降噪
语音识别:openai-whisper(内置), large-v3-turbo, 发音语言: 英语, 断句方式:LLM重新断句
翻译渠道:DeepSeek,原始语言:英语,目标语言:简体中文, 已选 发送完整字幕
输出字幕格式: 单字幕
代理地址:http://127.0.0.1:7897


=
system:Windows-10-10.0.19045-SP0
version:v4.11
frozen:True
language:zh
root_dir:H:/老骚/win-videotrans-v3.83

1 Replies

#1 DeepSeek reply at 1 hour ago

openai-whisper 内部依赖的 numba/llvmlite 版本不兼容导致子进程崩溃,常见于预打包版被旧补丁包覆盖后依赖错乱。

解决方案

  1. 改用 faster-whisper(内置) 渠道(推荐):在主界面语音识别渠道下拉框中切换为 faster-whisper(内置),模型仍选 large-v3-turbo。该渠道更稳定,无需额外修复。
  2. 重新下载完整包:若必须用 openai-whisper,请从官网重新下载 2.7GB 完整版压缩包,全量解压后再覆盖最新补丁包,避免依赖版本混乱。
  3. 源码部署用户:在虚拟环境中执行 uv sync 重新同步依赖,或手动升级 numballvmlite 至兼容版本:

    uv pip install --upgrade numba llvmlite

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