对于mkv格式的视频系统一直提示转录失败

[wzp006]

以下是两次运行的日志：

===========转录任务开始===========
时间：2025-05-21 22:36:34.653356
开始转换音频
转换为音频执行命令: ffmpeg -i E:\3333\JUR-032m1.wav -map 0:a -ac 1 -ar 16000 -af aresample=async=1 -y C:\Users\sumei\AppData\Local\Temp\tmpnt6zgo1v.wav
开始语音转录
Faster Whisper 执行命令: faster-whisper-xxl -m large-v3 --print_progress --model_dir E:\zimu\VideoCaptioner\AppData\models C:\Users\sumei\AppData\Local\Temp\bk_asr\tmpyv3c8su5\audio.wav -l ja -d cuda --output_format srt -o source --vad_filter true --vad_threshold 0.35 --vad_method silero_v4 --ff_mdx_kim2 --one_word 0 --sentence --max_line_width 30 --max_line_count 1 --max_comma 20 --max_comma_cent 50 --beep_off
Standalone Faster-Whisper-XXL r245.2 running on: CUDA
Note: 'large-v3' model may produce worse results than 'large-v2'!
Audio filtering is in progress...
2025-05-21 22:36:57 - faster_whisper - ERROR - Error: Some error in ffmpeg_audio() func.
Faster Whisper 返回值: 130
2025-05-21 22:36:57 - faster_whisper - ERROR - Faster Whisper 错误: Error: Some error in ffmpeg_audio() func.
2025-05-21 22:36:57 - transcript_thread - ERROR - 转录过程中发生错误: Error: Some error in ffmpeg_audio() func.
Traceback (most recent call last):
File "E:\zimu\VideoCaptioner\app\thread\transcript_thread.py", line 101, in run
asr_data = transcribe(
^^^^^^^^^^^
File "E:\zimu\VideoCaptioner\app\core\bk_asr\transcribe.py", line 86, in transcribe
asr_data = asr.run(callback=callback)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\zimu\VideoCaptioner\app\core\bk_asr\base.py", line 54, in run
resp_data = self._run(callback, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\zimu\VideoCaptioner\app\core\bk_asr\faster_whisper.py", line 266, in _run
raise RuntimeError(error_msg)
RuntimeError: Error: Some error in ffmpeg_audio() func.
Latest version info: v1.3.3
视频时长: 10836.12秒
2025-05-21 22:37:34 - video_info_thread - WARNING - 未找到视频流信息

===========转录任务开始===========
时间：2025-05-21 22:37:36.899990
开始转换音频
转换为音频执行命令: ffmpeg -i E:\3333\123.wav -map 0:a -ac 1 -ar 16000 -af aresample=async=1 -y C:\Users\sumei\AppData\Local\Temp\tmpbdi5na0l.wav
开始语音转录
Faster Whisper 执行命令: faster-whisper-xxl -m large-v3 --print_progress --model_dir E:\zimu\VideoCaptioner\AppData\models C:\Users\sumei\AppData\Local\Temp\bk_asr\tmp2cmeidrm\audio.wav -l ja -d cuda --output_format srt -o source --vad_filter true --vad_threshold 0.35 --vad_method silero_v4 --ff_mdx_kim2 --one_word 0 --sentence --max_line_width 30 --max_line_count 1 --max_comma 20 --max_comma_cent 50 --beep_off
Standalone Faster-Whisper-XXL r245.2 running on: CUDA
Note: 'large-v3' model may produce worse results than 'large-v2'!
Audio filtering is in progress...
2025-05-21 22:37:58 - faster_whisper - ERROR - Error: Some error in ffmpeg_audio() func.
Faster Whisper 返回值: 130
2025-05-21 22:37:59 - faster_whisper - ERROR - Faster Whisper 错误: Error: Some error in ffmpeg_audio() func.
2025-05-21 22:37:59 - transcript_thread - ERROR - 转录过程中发生错误: Error: Some error in ffmpeg_audio() func.
Traceback (most recent call last):
File "E:\zimu\VideoCaptioner\app\thread\transcript_thread.py", line 101, in run
asr_data = transcribe(
^^^^^^^^^^^
File "E:\zimu\VideoCaptioner\app\core\bk_asr\transcribe.py", line 86, in transcribe
asr_data = asr.run(callback=callback)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\zimu\VideoCaptioner\app\core\bk_asr\base.py", line 54, in run
resp_data = self._run(callback, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\zimu\VideoCaptioner\app\core\bk_asr\faster_whisper.py", line 266, in _run
raise RuntimeError(error_msg)
RuntimeError: Error: Some error in ffmpeg_audio() func.
Latest version info: v1.3.3

[caeserwl]

我也遇到了，不知道怎么处理，只能用剪映先变成mp4再转换

[philpw99]

可能是FasterWhisper里面“去除人声”的功能出错了，关掉这功能再试试