#7057 Audio listen unavailable for original voice clone

This error means pyVideoTrans cannot access/read the audio from the original video, which is required for clone voice cloning (it needs to extract reference audio slices from the original file).

Common root causes & fixes:

1. File path contains unsupported characters
   Move the video to a shallow, all-English/numeric path (e.g. D:/video.mp4). Avoid Chinese, spaces, or special symbols in the file name or the parent folders.
2. The video has no audio track, or the audio is corrupted/unreadable
   Play the video locally to confirm there is audible speech. If the video uses an unsupported codec (e.g. AV1), convert it to standard H.264/MP4 with FFmpeg first.
3. The file is locked by another process
   Close any media player, editor, or another program that may be holding the file. Also, don't start multiple sp.exe instances concurrently.
4. Output directory / tmp folder issue
   Do not click the 输出到... button to set a custom output directory when using external SRT/vocal/bgm files. Use the default _video_out structure.

5. Improper clone-related ASR settings
   Go to 菜单 -> 工具 -> 高级选项 -> 语音识别参数 and set:

   • Maximum speech duration: 6–10 seconds
   • Minimum speech duration: 3000–4000 ms
   • Check 合并过短字幕到邻近
   • Check Whisper预分割音频?

If the issue persists, provide the latest log file from D:/win-pyvideotrans-v4.10/logs/ (copy the last ~30 lines of error output) so we can identify the exact cause.

请查阅相关文档：

• 语音克隆 / 多角色配音
• 常见问题FAQ