语音识别阶段出错[Qwen-ASR(内置)] Qwen3ASREncoder expects padded_feature_length to be a multiple of n_window * 2 (100), but got 1315.:Traceback (most recent call last):
File "videotrans\process\stt_qwen.py", line 83, in qwen3asr_fun
File "D:\Translator\win-pyvideotrans-v4.16\_internal\torch\utils\_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)File "D:\Translator\win-pyvideotrans-v4.16\_internal\transformers\generation\utils.py", line 2645, in generate
result = decoding_method(File "D:\Translator\win-pyvideotrans-v4.16\_internal\transformers\generation\utils.py", line 2845, in _sample
outputs = self._prefill(File "D:\Translator\win-pyvideotrans-v4.16\_internal\transformers\generation\utils.py", line 3896, in _prefill
return self(**model_inputs, return_dict=True)File "D:\Translator\win-pyvideotrans-v4.16\_internal\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)Fil
......
udio_embeds = self.get_audio_features(input_features, input_features_mask, return_dict=True).pooler_output
File "D:\Translator\win-pyvideotrans-v4.16\_internal\transformers\utils\generic.py", line 911, in wrapper
output = func(self, *args, **kwargs)File "D:\Translator\win-pyvideotrans-v4.16\_internal\transformers\models\qwen3_asr\modeling_qwen3_asr.py", line 494, in get_audio_features
audio_output = self.audio_tower(File "D:\Translator\win-pyvideotrans-v4.16\_internal\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)File "D:\Translator\win-pyvideotrans-v4.16\_internal\torch\nn\modules\module.py", line 1762, in _call_impl
return forward_call(*args, **kwargs)File "D:\Translator\win-pyvideotrans-v4.16\_internal\transformers\utils\output_capturing.py", line 252, in wrapper
outputs = func(self, *args, **kwargs)File "D:\Translator\win-pyvideotrans-v4.16\_internal\transformers\models\qwen3_asr\modeling_qwen3_asr.py", line 383, in forward
raise ValueError(ValueError: Qwen3ASREncoder expects padded_feature_length to be a multiple of n_window * 2 (100), but got 1315.
trk=[TransCreate]: uuid='0bfae633bb', proxy_str='http://127.0.0.1:7897', cfg=[TaskCfgVTT]当前工作模式: 转录并翻译字幕
原始输入文件名: C:/Users/abc/Downloads/待翻译/test.m4a,
输出结果保存到文件夹: C:/Users/abc/Downloads/_video_out/待翻译/test-m4a,
临时文件夹: D:/Translator/win-pyvideotrans-v4.16/tmp/2900/0bfae633bb
已选 清理已存在
已启用CUDA加速
未选 降噪
语音识别:Qwen-ASR(内置), 1.7B, 发音语言: 英语,
翻译渠道:本地大模型(兼容OpenAI),原始语言:英语,目标语言:简体中文, 已选 发送完整字幕
输出字幕格式: 单字幕
代理地址: http://127.0.0.1:7897, precent=4, hasend=False, should_recogn=True, should_trans=True, should_dubbing=False, should_hebing=False, is_copy_video=False, video_codec_num=264, ignore_align=False, is_audio_trans=True, clone_ref='', should_recogn2=False, do_diarize=False
=
system:Windows-10-10.0.26200-SP0
version:v4.16
frozen:True
language:zh_CN
root_dir:D:/Translator/win-pyvideotrans-v4.16