Open hijos opened 3 months ago
如题。
日志文件:
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: 0.8.0 ==========2024-08-01_15:11:21========== ==========Start==========
current computer language region-format: zh_CN language: zh
==========2024-08-01_15:12:23========== ==========LoadModel==========
-model_size_or_path: large-v3 -device: cuda -device_index: 0 -compute_type: int8 -cpu_threads: 8 -num_workers: 1 -download_root: C:/Users/xxxxx/.cache/huggingface/hub -local_files_only: False -use_v3_model: True
Load over large-v3 max_length: 448 num_samples_per_token: 320 time_precision: 0.02 tokens_per_second: 50 input_stride: 2
[Using V3 model, modify number of mel-filters to 128]
==========2024-08-01_15:12:37========== ==========Process==========
redirect std output vad_filter : True -threshold : 0.5 -min_speech_duration_ms : 250 -max_speech_duration_s : inf -min_silence_duration_ms : 2000 -window_size_samples : 1024 -speech_pad_ms : 400 Transcribes options: -audio : ['C:/Users/xxxxx/Downloads/291441808_nb2-1-30280.m4s'] -language : zh -task : False -beam_size : 5 -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 : 2.4 -log_prob_threshold : -1.0 -no_speech_threshold : 0.6 -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 : "'.。,,!!??::”)]}、 -repetition_penalty : 1.0 -no_repeat_ngram_size : 0 -prompt_reset_on_temperature : 0.5 -max_new_tokens : None -chunk_length : 30.0 -clip_mode : 0 -clip_timestamps : 0 -hallucination_silence_threshold : 0.5 -hotwords : -language_detection_threshold : None -language_detection_segments : 1 create transcribe process with 1 workers start transcribe process
设置中已经选择了显卡
更新显卡驱动 系统设置里面还有个 GPU 加速选项检查试试看?
显卡驱动已经更新到最新了。 GPU加速选项也一直是开的。
如题。
日志文件:
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: 0.8.0 ==========2024-08-01_15:11:21========== ==========Start==========
current computer language region-format: zh_CN language: zh
==========2024-08-01_15:12:23========== ==========LoadModel==========
Load over large-v3 max_length: 448 num_samples_per_token: 320 time_precision: 0.02 tokens_per_second: 50 input_stride: 2
[Using V3 model, modify number of mel-filters to 128]
==========2024-08-01_15:12:37========== ==========Process==========
redirect std output vad_filter : True -threshold : 0.5 -min_speech_duration_ms : 250 -max_speech_duration_s : inf -min_silence_duration_ms : 2000 -window_size_samples : 1024 -speech_pad_ms : 400 Transcribes options: -audio : ['C:/Users/xxxxx/Downloads/291441808_nb2-1-30280.m4s'] -language : zh -task : False -beam_size : 5 -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 : 2.4 -log_prob_threshold : -1.0 -no_speech_threshold : 0.6 -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 : "'.。,,!!??::”)]}、 -repetition_penalty : 1.0 -no_repeat_ngram_size : 0 -prompt_reset_on_temperature : 0.5 -max_new_tokens : None -chunk_length : 30.0 -clip_mode : 0 -clip_timestamps : 0 -hallucination_silence_threshold : 0.5 -hotwords : -language_detection_threshold : None -language_detection_segments : 1 create transcribe process with 1 workers start transcribe process
设置中已经选择了显卡