#3981 videotrans.configure._except.StopRetry: D:/win-pyvideotrans-v3.98-0406/f5-tts/gire.WAV is not exists

216.227* Posted at: 8 hours ago 👁14

D:/win-pyvideotrans-v3.98-0406/f5-tts/gire.WAV is not exists
Traceback (most recent call last):
File "videotrans\task\_only_one.py", line 86, in run
File "videotrans\task\trans_create.py", line 715, in dubbing
File "videotrans\task\trans_create.py", line 993, in _tts
File "videotrans\tts\__init__.py", line 362, in run
File "videotrans\tts\_base.py", line 137, in run
File "videotrans\tts\_cosyvoice.py", line 26, in _exec
File "videotrans\tts\_base.py", line 210, in _local_mul_thread
File "videotrans\tts\_cosyvoice.py", line 157, in _item_task
File "videotrans\tts\_cosyvoice.py", line 54, in _item_task_cosyvoice2
videotrans.configure._except.StopRetry: D:/win-pyvideotrans-v3.98-0406/f5-tts/gire.WAV is not exists


=
system:Windows-10-10.0.19045-SP0
version:v3.98
frozen:True
language:zh
root_dir:D:/win-pyvideotrans-v3.98-0406

1 Replies

这是由于 CosyVoice 配音渠道在尝试使用声音克隆时,找不到指定的参考音频文件导致的。

解决方案:

  1. 确保你选择了 clone 配音角色,并且原视频语音清晰。
  2. 菜单 -> 工具 -> 高级选项 -> 语音识别参数 中,进行以下关键设置:

    • 最短语音持续毫秒:设置为 3000(确保参考音频时长不低于3秒)。
    • 最长语音持续秒数:设置为 10(确保参考音频时长不超过10秒)。
    • Whisper预分割音频?勾选此项。
    • 合并过短字幕到邻近勾选此项。
  3. 同时,在 设置更多参数... 中,勾选 分离人声背景声 选项,以提供更干净的参考音频。
  4. 完成以上设置后,重新处理视频。

原因说明:
当使用 clone 角色进行声音克隆时,软件会自动将每条字幕对应的原始语音片段作为参考音频发送给 TTS 服务(如 CosyVoice/F5-TTS)。这些服务通常要求参考音频时长在 3-10 秒之间。上述设置是为了确保生成的每条字幕时长能落在这个范围内,从而避免因参考音频文件不存在或不符合要求而报错。

请查阅相关文档:

Post Your Reply
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.

Related resource

Similar issues already exist