#8384 mkl_malloc: failed to allocate memory,Traceback (most recent call last): File

2405**afc7 Posted at: 1 hour ago

mkl_malloc: failed to allocate memory,Traceback (most recent call last):
File "videotrans\process\stt_faster.py", line 54, in _create_model
File "faster_whisper\transcribe.py", line 689, in init
RuntimeError: mkl_malloc: failed to allocate memory

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "videotrans\process\stt_faster.py", line 54, in _create_model
File "faster_whisper\transcribe.py", line 689, in init
RuntimeError: mkl_malloc: failed to allocate memory

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "videotrans\process\stt_faster.py", line 84, in faster_whisper
File "videotrans\process\stt_faster.py", line 73, in _create_model
File "videotrans\process\stt_faster.py", line 77, in _create_model
File "videotrans\process\stt_faster.py", line 54, in _create_model
File "faster_whisper\transcribe.py", line 689, in __ini
......
anscribe.py", line 689, in init
RuntimeError: mkl_malloc: failed to allocate memory

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "videotrans\process\stt_faster.py", line 54, in _create_model
File "faster_whisper\transcribe.py", line 689, in init
RuntimeError: mkl_malloc: failed to allocate memory

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "videotrans\process\stt_faster.py", line 84, in faster_whisper
File "videotrans\process\stt_faster.py", line 73, in _create_model
File "videotrans\process\stt_faster.py", line 77, in _create_model
File "videotrans\process\stt_faster.py", line 54, in _create_model
File "faster_whisper\transcribe.py", line 689, in init
RuntimeError: mkl_malloc: failed to allocate memory

trk=[TransCreate]: uuid='0008be8312', 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: D:/TAM MUOI CHAN HOA TAU/【ZTN】詹唐宁-三昧御光三昧真火/3.2.音光法门：DNA激活疗愈.mp4,
输出结果保存到文件夹: D:/TAM MUOI CHAN HOA TAU/【ZTN】詹唐宁-三昧御光三昧真火/_video_out/3.2.音光法门：DNA激活疗愈-mp4,
临时文件夹: C:/Users/hoana/Downloads/win-pyvideotrans-v4.16/tmp/14048/0008be8312
已选清理已存在
未启用CUDA加速
未选降噪
STT Service:faster-whisper(Built-in), medium, 发音语言: Simplified Chinese,
TranslateSrt:Google Translate (Free),原始语言:Simplified Chinese,目标语言:Vietnamese, 未选 SendSRT
TTS Service:Edge-TTS (Free), 角色:HoaiMy(Female/VN), 配音语言:Vietnamese, 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
已选移除每条字幕配音开头和结尾静音缓冲
静音移除力度: default
字幕: No Subtitles , precent=4, hasend=False, should_recogn=True, should_trans=True, should_dubbing=True, 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=9058960

=
system:Windows-10-10.0.26200-SP0
version:v4.16
frozen:True
language:en_US
root_dir:C:/Users/hoana/Downloads/win-pyvideotrans-v4.16

#1 DeepSeek reply at 1 hour ago

This indicates the system (RAM, or GPU VRAM if CUDA were enabled) cannot allocate enough memory to load the faster-whisper medium model. Your log shows 未启用CUDA加速 (CUDA acceleration NOT enabled), meaning the model is loading entirely into system RAM — and medium requires roughly 5 GB free, which your machine apparently does not have available at the moment of loading.

Solution (apply in order, stop once it works):

Reduce model size. In the main window 语音识别(STT Service) dropdown, change the model from medium to small or base. base needs ~1 GB, small ~2 GB. This is the fastest fix.
Free up RAM before starting. Close browsers, other apps; or reboot. Verify you have at least 5–8 GB free RAM. This mkl_malloc error is often triggered simply because the machine was low on available memory.
Enable CUDA if you have an NVIDIA GPU. If you do, tick CUDA加速?(Enable CUDA?) in the main window — this runs the model on VRAM instead of RAM and is also several times faster. Requires CUDA Toolkit 12.8+ and cuDNN 9.11+ installed and in PATH.
Lower compute precision. Go to Menu → Tools/Options → Advanced Options → 语音识别参数(ASR Settings), set 计算数据类型(Compute data type) to int8 (least memory) — and set 识别准确度beam_size and best_of to 1 to further reduce peak memory.
Ensure GPU同时任务数 = 1 in Advanced Options → 通用设置, and check whether another translation task is still holding memory in the background.

Note: medium on CPU also means very slow processing — if you have an NVIDIA GPU, enabling CUDA is strongly recommended.

#8384 mkl_malloc: failed to allocate memory,Traceback (most recent call last): File

1 Replies

Post Your Reply

Related resource