0.8.5版本加载不运行

[11123kbbs]

模型加载了。但是卡在这不运行了。请问怎么处理。低版本的是可以正常运行的。

[11123kbbs]

[CheshireCC]

看看软件日志，应该是显存不够了

[11123kbbs]

4090为什么会显存不够。我使用3.1.1版本是正常的

[CheshireCC]

4090为什么会显存不够。我使用3.1.1版本是正常的

看看日志

[11123kbbs]

2024-12-27_01:22:58 - faster_whisper - INFO - Processing audio with duration 30:00.000
2024-12-27_01:23:01 - faster_whisper - INFO - VAD filter removed 00:54.192 of audio
2024-12-27_01:23:01 - faster_whisper - DEBUG - VAD filter kept the following audio segments: [00:00.000 -> 00:01.520], [00:05.616 -> 05:51.120], [05:52.752 -> 06:55.920], [06:58.608 -> 17:49.616], [17:50.960 -> 17:56.560], [17:58.224 -> 20:00.784], [20:05.904 -> 20:09.584], [20:12.720 -> 20:15.952], [20:20.112 -> 20:22.032], [20:24.624 -> 21:45.616], [21:47.056 -> 23:05.712], [23:11.216 -> 23:13.296], [23:16.304 -> 23:18.224], [23:21.008 -> 25:16.560], [25:17.872 -> 27:48.656], [27:50.576 -> 29:22.544], [29:26.736 -> 29:44.336], [29:46.992 -> 29:55.056]
2024-12-27_01:23:01 - faster_whisper - DEBUG - Processing segment at 00:00.000

这个就是日志所有显示内容

[CheshireCC]

另一个日志

[11123kbbs]

The torchaudio backend is switched to 'soundfile'. Note that 'sox_io' is not supported on Windows.
torchvision is not available - cannot save figures
The torchaudio backend is switched to 'soundfile'. Note that 'sox_io' is not supported on Windows.

faster_whisper_GUI: 0.8.5
==========2024-12-28_16:46:15==========
==========Start==========

language: zh

==========2024-12-28_16:46:27==========
==========LoadModel==========

-model_size_or_path: F:/whisper-large-v3-float32
-device: cuda
-device_index: 0
-compute_type: float32
-cpu_threads: 4
-num_workers: 1
-download_root: C:/Users/ab695/.cache/huggingface/hub
-local_files_only: False
-use_v3_model: False

Load over
F:/whisper-large-v3-float32
max_length: 448
num_samples_per_token: 320
time_precision: 0.02
tokens_per_second: 50
input_stride: 2

==========2024-12-28_16:46:45==========
==========Process==========

redirect std output
vad_filter : True
-onset : 0.2
-min_speech_duration_ms : 0
-max_speech_duration_s : inf
-min_silence_duration_ms : 2000
-speech_pad_ms : 400
Transcribes options:
-audio : ['C:/Users/ab695/Downloads/新建文件夹/RAW VIDEO/TUTO_001_Car_Preparation.mp4']
-language : en
-task : False
-log_progress : False
-beam_size : 1
-best_of : 5
-patience : 1.0
-length_penalty : 1.0
-temperature : [0.0, 0.2, 0.4, 0.6, 0.8, 1.0]
-compression_ratio_threshold : 1.4
-log_prob_threshold : -10.0
-no_speech_threshold : 0.9
-condition_on_previous_text : False
-initial_prompt : None
-prefix : None
-suppress_blank : True
-suppress_tokens : [-1]
-without_timestamps : False
-max_initial_timestamp : 1.0
-word_timestamps : False
-prepend_punctuations : "'“¿([{-
-append_punctuations : "'.。,，!！?？:：”)]}、
-multilingual : False
-repetition_penalty : 1.0
-no_repeat_ngram_size : 0
-prompt_reset_on_temperature : 0.5
-max_new_tokens : None
-chunk_length : 30
-clip_mode : 0
-clip_timestamps : 0
-hallucination_silence_threshold : 0.5
-hotwords :
-language_detection_threshold : 0.5
-language_detection_segments : 1
create transcribe process with 1 workers
start transcribe process
Traceback (most recent call last):
File "C:\Users\ab695\Desktop\新建文1\faster_whisper_GUI\transcribe.py", line 371, in run
File "C:\Users\ab695\Desktop\新建文1\concurrent\futures_base.py", line 621, in result_iterator
File "C:\Users\ab695\Desktop\新建文1\concurrent\futures_base.py", line 319, in _result_or_cancel
File "C:\Users\ab695\Desktop\新建文1\concurrent\futures_base.py", line 458, in result
File "C:\Users\ab695\Desktop\新建文1\concurrent\futures_base.py", line 403, in __get_result
File "C:\Users\ab695\Desktop\新建文1\concurrent\futures\thread.py", line 58, in run
File "C:\Users\ab695\Desktop\新建文1\faster_whisper_GUI\transcribe.py", line 281, in transcribe_file
File "C:\Users\ab695\Desktop\新建文1\faster_whisper\transcribe.py", line 1800, in restore_speech_timestamps
File "C:\Users\ab695\Desktop\新建文1\faster_whisper\transcribe.py", line 1138, in generate_segments
File "C:\Users\ab695\Desktop\新建文1\faster_whisper\transcribe.py", line 1348, in encode
ValueError: Invalid input features shape: expected an input with shape (1, 128, 3000), but got an input with shape (1, 80, 3000) instead

[CheshireCC]

V3 模型选项问题

[11123kbbs]

你好，找个模型选项问题怎么修改呀。可以告诉下吗？谢谢

[weituo2002]

同问，也是卡住了不知道该怎么修改参数

[11123kbbs]

作者，这个怎么修改或者选择呀。谢谢了

[CheshireCC]

加载模型页面可以设置