#5683 CUDA error: device-side assert triggeredCompile with TORCH_USE_CUDA_DSA to enable device-side assertions.,Traceback (mos

1.204* Posted at: 1 hour ago

CUDA error: device-side assert triggered
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
,Traceback (most recent call last):
File "videotrans\process\tts_fun.py", line 127, in qwen3tts_fun
File "torch\utils\_contextlib.py", line 116, in decorate_context

return func(*args, **kwargs)

File "D:\AIGC\pyVideoTrans\_internal\qwen_tts\inference\qwen3_tts_model.py", line 603, in generate_voice_clone

talker_codes_list, _ = self.model.generate(

File "torch\utils\_contextlib.py", line 116, in decorate_context

return func(*args, **kwargs)

File "D:\AIGC\pyVideoTrans\_internal\qwen_tts\core\models\modeling_qwen3_tts.py", line 2272, in generate

talker_result = self.talker.generate(

File "torch\utils\_contextlib.py", line 116, in decorate_context

return func(*args, **kwargs)

File "transformers\generation\utils.py", line 2566, in generate

result = decoding_method(

File "transformers\generation\utils.py", line 2789, in _sample

......

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

File "torch\nn\modules\module.py", line 1762, in _call_impl

return forward_call(*args, **kwargs)

File "transformers\utils\generic.py", line 918, in wrapper

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

File "D:\AIGC\pyVideoTrans\_internal\qwen_tts\core\models\modeling_qwen3_tts.py", line 1671, in forward

predictor_result = self.code_predictor.generate(

File "torch\utils\_contextlib.py", line 116, in decorate_context

return func(*args, **kwargs)

File "transformers\generation\utils.py", line 2566, in generate

result = decoding_method(

File "transformers\generation\utils.py", line 2831, in _sample

next_tokens = torch.multinomial(probs, num_samples=1).squeeze(1)

RuntimeError: CUDA error: device-side assert triggered
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

Traceback (most recent call last):
File "videotrans\task\only_one.py", line 86, in run
File "videotrans\task\_stage_dubbing.py", line 23, in dubbing
File "videotrans\task\_stage_dubbing.py", line 102, in _tts
File "videotrans\tts\__init__.py", line 207, in run
File "videotrans\tts\_base.py", line 93, in run
File "videotrans\tts\_qwenttslocal.py", line 46, in _exec
File "videotrans\configure\base.py", line 272, in _new_process
videotrans.configure.excepts.VideoTransError: CUDA error: device-side assert triggered
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
,Traceback (most recent call last):
File "videotrans\process\tts_fun.py", line 127, in qwen3tts_fun
File "torch\utils\_contextlib.py", line 116, in decorate_context

return func(*args, **kwargs)

File "D:\AIGC\pyVideoTrans\_internal\qwen_tts\inference\qwen3_tts_model.py", line 603, in generate_voice_clone

talker_codes_list, _ = self.model.generate(

File "torch\utils\_contextlib.py", line 116, in decorate_context

return func(*args, **kwargs)

File "D:\AIGC\pyVideoTrans\_internal\qwen_tts\core\models\modeling_qwen3_tts.py", line 2272, in generate

talker_result = self.talker.generate(

File "torch\utils\_contextlib.py", line 116, in decorate_context

return func(*args, **kwargs)

File "transformers\generation\utils.py", line 2566, in generate

result = decoding_method(

File "transformers\generation\utils.py", line 2789, in _sample

outputs = model_forward(**model_inputs, return_dict=True)

File "torch\nn\modules\module.py", line 1751, in _wrapped_call_impl

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

File "torch\nn\modules\module.py", line 1762, in _call_impl

return forward_call(*args, **kwargs)

File "transformers\utils\generic.py", line 918, in wrapper

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

File "D:\AIGC\pyVideoTrans\_internal\qwen_tts\core\models\modeling_qwen3_tts.py", line 1671, in forward

predictor_result = self.code_predictor.generate(

File "torch\utils\_contextlib.py", line 116, in decorate_context

return func(*args, **kwargs)

File "transformers\generation\utils.py", line 2566, in generate

result = decoding_method(

File "transformers\generation\utils.py", line 2831, in _sample

next_tokens = torch.multinomial(probs, num_samples=1).squeeze(1)

RuntimeError: CUDA error: device-side assert triggered
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

cfg=[TaskCfgVTT]当前工作模式: 翻译视频或音频 单视频模式
原始输入文件名: D:/Bandicam/ZWEI2PDX9 2026-07-08 00-43-34-057.mp4,
输出结果保存到文件夹: D:/Bandicam/_video_out/ZWEI2PDX9 2026-07-08 00-43-34-057-mp4,
临时文件夹: D:/AIGC/pyVideoTrans/tmp/10964/482438eea5
已选 清理已存在
已选 启用CUDA加速
未选 降噪
语音识别:faster-whisper(本地内置), model_name: large-v3-turbo, 发音语言: 英语, 断句方式:默认断句
翻译渠道:DeepSeek,原始语言:英语,目标语言:简体中文, 已选 发送完整字幕
配音渠道:Qwen3-TTS(本地内置), 角色:clone, 配音语言:简体中文, 未选 二次语音识别
音量:+0%, 语速:+0%, 已选 音频加速, 未选 视频慢速
字幕: 嵌入硬字幕
已选 仅输出mp4
代理地址:


=
system:Windows-10-10.0.26200-SP0
version:v4.04
frozen:True
language:zh
root_dir:D:/AIGC/pyVideoTrans

1 Replies

这是由于 CUDA 显存不足或 GPU 计算环境异常导致的 device-side assert triggered 错误。当使用 Qwen3-TTS(本地内置) 并且角色设为 clone 时,模型需要额外加载参考音频并进行分身计算,加上你已经选用了 large-v3-turbo 模型,两者叠加消耗显存,导致 GPU 断言失败。

解决方法(按优先级排序):

  1. 降低计算数据类型
    进入 菜单 → 工具 → 高级选项 → 语音识别参数,将 计算数据类型float32 改为 float16int8(推荐 float16),保存后重启软件重试。
  2. 降低 ASR 模型大小
    在主界面将语音识别模型从 large-v3-turbo 改为 mediumsmall,释放部分显存给 TTS 使用。
  3. 临时关闭 CUDA 加速
    如果以上仍无效,取消勾选主界面上的 启用CUDA加速,改用 CPU 运行(速度会大幅降低,但可避免显存溢出)。
  4. 切换为其他配音渠道(推荐绕过)
    若不需要克隆,可将配音渠道从 Qwen3-TTS(本地内置) 改为 Edge-TTS(免费、在线、不消耗本地显存);若需要克隆,可尝试 F5-TTS(本地内置)OmniVoice-TTS(本地API),它们在同样显存下更稳定。
  5. 调整 clone 参考音频时长
    进入 菜单 → 工具 → 高级选项 → 语音识别参数,将 最长语音持续秒数 设为 6最短语音持续毫秒 设为 3000,并勾选 合并过短字幕到邻近,避免裁剪出过长的参考音频导致显存激增。

注意: 你的视频文件名包含空格和中文字符(ZWEI2PDX9 2026-07-08 00-43-34-057.mp4),建议重命名为纯英文(如 video1.mp4)再尝试,虽然当前错误不是路径问题,但也可以排除潜在干扰。

如果问题仍然存在,请提供 logs/ 目录下最新日志文件的底部 30 行。

Post Your Reply
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: ko-fi.com/jianchang512

Related resource