#8599 No human voice detected. Please confirm that human speech is present in the original file. [info.duration_after_vad=0.0s

103.151**1 Posted at: 1 hour ago

No human voice detected. Please confirm that human speech is present in the original file. [info.duration_after_vad=0.0s].
_kw={'beam_size': 5, 'best_of': 5, 'condition_on_previous_text': False, 'threshold': 0.45, 'no_speech_threshold': 0.6, 'temperature': 0.0, 'repetition_penalty': 1.0, 'compression_ratio_threshold': 2.4, 'initial_prompt': None}
info=TranscriptionInfo(language='en', language_probability=0.5171349048614502, duration=1235.4346875, duration_after_vad=0.0, all_language_probs=[('en', 0.5171349048614502), ('ja', 0.07975884526968002), ('ru', 0.06972990930080414), ('zh', 0.04287295043468475), ('es', 0.03771531581878662), ('ko', 0.03457534313201904), ('pl', 0.03138734772801399), ('de', 0.03044981323182583), ('pt', 0.02156747505068779), ('fr', 0.021338684484362602), ('it', 0.019429387524724007), ('tr', 0.015838202089071274), ('id', 0.014627281576395035), ('nl', 0.007678893394768238), ('cs', 0.006308021489530802), ('vi', 0.005704732611775398), ('hu', 0.004874781705439091), ('f
......
pression_ratio_threshold=2.4, condition_on_previous_text=False, prompt_reset_on_temperature=0.5, temperatures=[0.0], 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, 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=0.45, neg_threshold=None, min_speech_duration_ms=0, max_speech_duration_s=inf, min_silence_duration_ms=2000, speech_pad_ms=400))

trk=[TransCreate]: uuid='5a2fb5fa0d', proxy_str=None, cfg=[TaskCfgVTT]当前工作模式: 翻译视频 单视频模式
原始输入文件名: F:/sp2/407340-1080p.mp4,
输出结果保存到文件夹: F:/sp2/_video_out/407340-1080p-mp4,
临时文件夹: E:/win-pyvideotrans-v4.17-1028/tmp/23852/5a2fb5fa0d
已选 清理已存在
未启用CUDA加速
语音识别:faster-whisper(内置), large-v3-turbo, 发音语言: 自动检测, , 数据类型:default
翻译渠道:Google(免费),原始语言:自动检测,目标语言:简体中文, 未选 发送完整字幕, settings.get('trans_thread')=5, settings.get('aitrans_thread')=40
未选 配音角色,不进行配音
字幕: 嵌入硬字幕
已选 仅输出mp4, precent=4, hasend=False, should_recogn=True, should_trans=True, should_dubbing=False, should_hebing=True, is_copy_video=True, video_codec_num=264, ignore_align=False, is_audio_trans=False, clone_ref='', should_recogn2=False, do_diarize=True
video_duration_ms=1235434


=
system:Windows-10-10.0.22631-SP0
version:v4.17
frozen:True
language:zh_CN
root_dir:E:/win-pyvideotrans-v4.17-1028

1 Replies

#1 DeepSeek reply at 1 hour ago

duration_after_vad=0.0s 说明 VAD(语音活动检测)在整个 1235 秒音轨中未检测到任何符合阈值的语音片段,因此放行到 faster-whisper 前就已被拦截,并非模型识别失败。

排查与处理(按顺序尝试):

  1. 先验证音频本身:用播放器打开 F:/sp2/407340-1080p.mp4 听是否有清晰人声。若该视频只有背景音乐、纯音效或人声极轻,本软件确实无法处理(无法处理无人声说话的音频)。
  2. 调低 VAD 灵敏度阈值:进入 菜单 → 工具/选项 → 高级选项 → 语音识别参数,把 语音阈值 从默认 0.45 调低到 0.30 或 0.25(越小越灵敏,但可能把噪声误判为语音),同时把 非语音阈值 也适当调低,保存后重试。勾选主界面 清理已生成。
  3. 改善源音频质量:在 设置更多参数 中勾选 降噪;若背景音乐很强,勾选 分离人声背景声,再重试。
  4. 更换识别渠道做交叉验证:把语音识别渠道换成 openai-whisper(内置) 或 Qwen-ASR(内置) 再跑一次,以判断是否是单一渠道 VAD 判定问题。
  5. 确认发音语言:日志中 language_probability=0.517(en),概率偏低说明音频可能并非清晰人声。若确知语言,请手动指定 发音语言,不要用 自动检测。
  6. 确认视频确有音轨:日志显示 duration=1235s 说明提取到了音频流,但若是从某些网站分离下载的视频,音轨可能实际为静音或极低音量,可先用格式转换功能把音频单独导出验证。

若以上均无效,请提供 logs/ 目录下当日日志尾部约 30 行,以及该视频的音轨信息(可用 菜单 → 工具 → 视频与音频合并 之外的方式确认),以便进一步判断。

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