#7966 语音识别阶段出错[Nemotron-3.5-asr-0.6b(内置)] Sequence Length: 12463 has to be less or equal than config.max_position_embeddings 5

语音识别阶段出错[Nemotron-3.5-asr-0.6b(内置)] Sequence Length: 12463 has to be less or equal than config.max_position_embeddings 5000.:Traceback (most recent call last):
File "videotrans\process\stt_nemotron.py", line 41, in nemotron_asr
File "D:\program\videotrans\_internal\transformers\models\nemotron3_5_asr\generation_nemotron3_5_asr.py", line 35, in generate

return super().generate(inputs=inputs, generation_config=generation_config, **kwargs)

File "D:\program\videotrans\_internal\transformers\models\nemotron_asr_streaming\generation_nemotron_asr_streaming.py", line 228, in generate

outputs = super().generate(inputs=inputs, generation_config=generation_config, **kwargs)

File "D:\program\videotrans\_internal\transformers\models\parakeet\generation_parakeet.py", line 256, in generate

outputs = super().generate(inputs=inputs, generation_config=generation_config, **kwargs)

File "D:\program\videotrans\_internal\torch\utils\_contextlib.py", line 116, in decorate_contex
......
tput_capturing.py", line 252, in wrapper

outputs = func(self, *args, **kwargs)

File "D:\program\videotrans\_internal\transformers\utils\generic.py", line 911, in wrapper

output = func(self, *args, **kwargs)

File "D:\program\videotrans\_internal\transformers\models\nemotron_asr_streaming\modeling_nemotron_asr_streaming.py", line 1007, in forward

position_embeddings = self.encode_positions(inputs_embeds, cached_frames=cached_frames)

File "D:\program\videotrans\_internal\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl

return self._call_impl(*args, **kwargs)

File "D:\program\videotrans\_internal\torch\nn\modules\module.py", line 1762, in _call_impl

return forward_call(*args, **kwargs)

File "D:\program\videotrans\_internal\torch\utils\_contextlib.py", line 116, in decorate_context

return func(*args, **kwargs)

File "D:\program\videotrans\_internal\transformers\models\nemotron_asr_streaming\modeling_nemotron_asr_streaming.py", line 376, in forward

raise ValueError(

ValueError: Sequence Length: 12463 has to be less or equal than config.max_position_embeddings 5000.

self.cfg=[TaskCfgVTT]当前工作模式: 转录并翻译字幕
原始输入文件名: D:/Downloads/Videos/!Submaker/HUNTA-038-U_01.mp4,
输出结果保存到文件夹: D:/Downloads/Videos/!Submaker/HUNTA-038-U_01-mp4,
临时文件夹: D:/program/videotrans/tmp/14884/ad3d7e9ef9
已选 清理已存在
未启用CUDA加速
未选 降噪
语音识别:Nemotron-3.5-asr-0.6b(内置), , 发音语言: 日语, 断句方式:LLM重新断句
翻译渠道:DeepSeek,原始语言:日语,目标语言:简体中文, 未选 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