#8423 语音识别阶段出错[faster-whisper(本地内置)] No transcription results returned. Please check the original audio/video or model and try

125.67**4 Posted at: 56 minutes ago

语音识别阶段出错[faster-whisper(本地内置)] No transcription results returned. Please check the original audio/video or model and try again.
info=TranscriptionInfo(language='en', language_probability=1, duration=179.8153125, duration_after_vad=0.0, all_language_probs=None, transcription_options=TranscriptionOptions(beam_size=5, best_of=5, patience=1, length_penalty=1, repetition_penalty=1.0, no_repeat_ngram_size=0, log_prob_threshold=-1.0, no_speech_threshold=0.6, compression_ratio_threshold=2.4, condition_on_previous_text=False, prompt_reset_on_temperature=0.5, temperatures=[0.0, 0.2, 0.4, 0.6, 0.8, 1.0], initial_prompt=None, prefix=None, suppress_blank=True, suppress_tokens=(1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 359, 503, 522, 542, 873, 893, 902, 918, 922, 931, 1350, 1853, 1982, 2460, 2627, 3246, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 1655
......
, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 16553, 16604, 18362, 18956, 20075, 21675, 22520, 26130, 26161, 26435, 28279, 29464, 31650, 32302, 32470, 36865, 42863, 47425, 49870, 50254, 50258, 50359, 50360, 50361, 50362, 50363), without_timestamps=False, max_initial_timestamp=1.0, word_timestamps=True, prepend_punctuations='"\'“¿([{-', append_punctuations='"\'.。,,!!??::”)]}、', multilingual=False, max_new_tokens=None, clip_timestamps=[0.0], hallucination_silence_threshold=None, hotwords=''), vad_options=VadOptions(threshold=0.5, neg_threshold=None, min_speech_duration_ms=0, max_speech_duration_s=inf, min_silence_duration_ms=140, speech_pad_ms=400))
cfg=TaskCfgSTT(uuid='4e9e64b3b1', name='D:/BaiduNetdiskDownload/吃饱了就想玩/吃饱了就想玩/jjazzlab/03.JJazzLab_ 4 个实用小技巧_1080p.mp4', dirname='D:/BaiduNetdiskDownload/吃饱了就想玩/吃饱了就想玩/jjazzlab', noextname='03.JJazzLab_ 4 个实用小技巧_1080p', basename='03.JJazzLab_ 4 个实用小技巧_1080p.mp4', ext='mp4', target_dir='D:/BaiduNetdiskDownload/pyVideoTrans/win-pyvideotrans-v4.07/output/recogn', cache_folder='D:/BaiduNetdiskDownload/pyVideoTrans/win-pyvideotrans-v4.07/tmp/29788/4e9e64b3b1', is_cuda=True, source_language=None, source_language_code=None, source_sub=None, source_wav=None, source_wav_output=None, target_language=None, target_language_code=None, target_sub='D:/BaiduNetdiskDownload/pyVideoTrans/win-pyvideotrans-v4.07/output/recogn/03.JJazzLab_ 4 个实用小技巧_1080p.srt', target_wav=None, target_wav_output=None, detect_language='en', recogn_type=0, model_name='large-v3-turbo', shibie_audio='D:/BaiduNetdiskDownload/pyVideoTrans/win-pyvideotrans-v4.07/tmp/29788/4e9e64b3b1/03.JJazzLab_ 4 个实用小技巧_1080p-1792812283.9391325.wav', remove_noise=False, enable_diariz=False, nums_diariz=0, rephrase=0, fix_punc=0)


=
system:Windows-10-10.0.26100-SP0
version:v4.07
frozen:True
language:zh
root_dir:D:/BaiduNetdiskDownload/pyVideoTrans/win-pyvideotrans-v4.07

1 Replies

#1 DeepSeek reply at 55 minutes ago

duration_after_vad=0.0 表示 VAD 语音活动检测把整段 179 秒音频全部判定为静音,没有切出任何语音片段送给模型,因此 faster-whisper 返回 0 条结果。这不是模型损坏或下载失败,而是 VAD 没检测到“人声”。

排查与解决步骤

  1. 先确认视频真的有人声
    用播放器打开该 mp4 听一下,确认不是纯音乐/纯演示操作/音量极小的视频。若是纯 BGM 无解说,本软件无法处理。
  2. 调低 VAD 语音阈值(最常见解法)
    菜单 → 工具/选项 → 高级选项 → 【语音识别参数】:

    • 语音阈值:默认 0.45,改为 0.2 ~ 0.3
    • 非语音阈值:适当调低
      保存后,勾选主界面 清理已生成,重新执行。
  3. 尝试切换 VAD 类型
    同一区域内 选择VAD:默认 silero,可切换为 tenvad(或反之)再试,两者对低音量人声的灵敏度不同。
  4. 放宽切分参数

    • 静音分割持续毫秒:从 140 调大到 300~500
    • 最短语音持续(毫秒):调低到 500~1000
      避免短促人声被当成静音过滤掉。
  5. 若原音频音量很小或有明显底噪
    在 设置更多参数 中勾选 降噪;或勾选 分离人声背景声 后再识别。
  6. 交叉验证
    用左侧面板 语音转录 功能单独转录这个视频,发音语言 选 自动检测,若同样返回空,则基本确认为音频本身无人声或音量过低。

补充

  • 你的版本是 v4.07,较旧,建议升级到最新版(VAD 切分算法在 v4.12 之后有优化)。
  • 路径 D:/BaiduNetdiskDownload/吃饱了就想玩/.../03.JJazzLab_ 4 个实用小技巧_1080p.mp4 含中文和空格,虽非本次直接原因,仍建议改为纯英文短路径(如 D:/videos/03.mp4)以避免后续 ffmpeg 相关报错。

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