#7198 配音阶段出错 [Qwen3-TTS(内置)] Error no file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt.index or flax_m

168.158**6 Posted at: 2 hours ago

配音阶段出错 [Qwen3-TTS(内置)] Error no file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt.index or flax_model.msgpack found in directory D:/sougou/win-pyvideotrans-v4.112/models/models--Qwen--Qwen3-TTS-12Hz-0.6B-CustomVoice.,Traceback (most recent call last):
File "videotrans\process\qwen_tts.py", line 54, in qwen3tts_fun
File "qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained

model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs)

File "D:\sougou\win-pyvideotrans-v4.112\_internal\transformers4576\models\auto\auto_factory.py", line 604, in from_pretrained

return model_class.from_pretrained(

File "qwen_tts\core\models\modeling_qwen3_tts.py", line 1876, in from_pretrained

model = super().from_pretrained(

File "D:\sougou\win-pyvideotrans-v4.112\_internal\transformers4576\modeling_utils.py", line 277, in _wrapper

return func(*args, **kwargs)

File "D:\sougou\win-pyvideotrans-v4.112\_internal\transform
......
ile named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt.index or flax_model.msgpack found in directory D:/sougou/win-pyvideotrans-v4.112/models/models--Qwen--Qwen3-TTS-12Hz-0.6B-CustomVoice.,Traceback (most recent call last):
File "videotrans\process\qwen_tts.py", line 54, in qwen3tts_fun
File "qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained

model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs)

File "D:\sougou\win-pyvideotrans-v4.112\_internal\transformers4576\models\auto\auto_factory.py", line 604, in from_pretrained

return model_class.from_pretrained(

File "qwen_tts\core\models\modeling_qwen3_tts.py", line 1876, in from_pretrained

model = super().from_pretrained(

File "D:\sougou\win-pyvideotrans-v4.112\_internal\transformers4576\modeling_utils.py", line 277, in _wrapper

return func(*args, **kwargs)

File "D:\sougou\win-pyvideotrans-v4.112\_internal\transformers4576\modeling_utils.py", line 4896, in from_pretrained

checkpoint_files, sharded_metadata = _get_resolved_checkpoint_files(

File "D:\sougou\win-pyvideotrans-v4.112\_internal\transformers4576\modeling_utils.py", line 989, in _get_resolved_checkpoint_files

raise OSError(

OSError: Error no file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt.index or flax_model.msgpack found in directory D:/sougou/win-pyvideotrans-v4.112/models/models--Qwen--Qwen3-TTS-12Hz-0.6B-CustomVoice.
[DubbingSrt]语音合成: self.out_ext='wav',self.is_multi_role=True
[TaskCfgTTS]
原始输入文件名: D:/win-pyvideotrans-v4.11 2/output/recogn/9月10日.srt,
输出结果保存到文件夹: D:/搜狗高速下载/win-pyvideotrans-v4.11 2/output/tts,
临时文件夹: D:/sougou/win-pyvideotrans-v4.112/tmp/19892/d11c74b8e7
未启用CUDA加速
配音渠道:Qwen3-TTS(内置), 角色:No, 配音语言:en
音量:+100%, 语速:+0%, 未选 音频加速
未选 移除字幕间空隙, 未选 强制对齐字幕和声音
代理地址:http://127.0.0.1:10809


=
system:Windows-10-10.0.26200-SP0
version:v4.11
frozen:True
language:zh
root_dir:D:/sougou/win-pyvideotrans-v4.112

1 Replies

#1 DeepSeek reply at 2 hours ago

核心原因:Qwen3-TTS(内置) 模型文件未下载完整或下载失败。你目录 models/models--Qwen--Qwen3-TTS-12Hz-0.6B-CustomVoice 内缺少模型权重文件(model.safetensors 等),因此加载时报此错。

处理步骤:

  1. 删除残缺模型目录:彻底删除
    D:/sougou/win-pyvideotrans-v4.112/models/models--Qwen--Qwen3-TTS-12Hz-0.6B-CustomVoice
    文件夹(下载不完整时,目录看起来存在但实际缺少权重文件)。
  2. 手动下载模型(自动下载常因网络失败):

    • CustomVoice 模型:
      https://huggingface.co/Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice/tree/main
    • 同时建议一并下载 Base 模型:
      https://huggingface.co/Qwen/Qwen3-TTS-12Hz-0.6B-Base/tree/main
  3. 放置位置

    • 软件目录/models/models--Qwen--Qwen3-TTS-12Hz-0.6B-CustomVoice
    • 软件目录/models/models--Qwen--Qwen3-TTS-12Hz-0.6B-Base
      把你下载的全部文件放入对应目录(注意:若浏览器自动重命名了同名文件如 model(2).safetensors,必须改回原名)。
  4. 放置完成后重启软件,重新执行配音任务。

补充提示:

  • 你的日志显示 未启用CUDA加速。Qwen3-TTS 属"优先使用GPU"渠道,若你有 N 卡,建议配置 CUDA 12.8 + cuDNN 9.11 并勾选加速,否则纯 CPU 运行很慢。
  • 备用方案:若手动下载仍困难,可临时改用 Edge-TTS(免费) 渠道配音。

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