Faster Whisper转录失败，请求帮助

[eytion]

配置：RTX5080 驱动：595.71

LOG如下：
开始语音转录
Faster Whisper 执行命令: faster-whisper-xxl -m tiny --print_progress --model_dir D:\Program Files\VideoCaptioner\AppData\models C:\Users\zhuyi\AppData\Local\Temp\bk_asr\tmpawi_yoh8\audio.wav -l ja -d cuda --output_format srt -o source --vad_filter true --vad_threshold 0.40 --vad_method silero_v4 --one_word 0 --sentence --max_line_width 30 --max_line_count 1 --max_comma 20 --max_comma_cent 50 --beep_off
Model not found at: D:\Program Files\VideoCaptioner\AppData\models\faster-whisper-tiny
Attempting to download:
tokenizer.json: 0.00B [00:00, ?B/s]
config.json: 0.00B [00:00, ?B/s]�[A
config.json: 2.25kB [00:00, 4.47MB/s]
vocabulary.txt: 0.00B [00:00, ?B/s]�[A
tokenizer.json: 901kB [00:00, 1.60MB/s]
tokenizer.json: 1.34MB [00:00, 1.84MB/s]
tokenizer.json: 2.20MB [00:00, 3.00MB/s]
tokenizer.json: 2.20MB [00:00, 2.50MB/s]
vocabulary.txt: 460kB [00:00, 827kB/s]�[A
vocabulary.txt: 460kB [00:00, 826kB/s]
model.bin: 0%| | 0.00/75.5M [00:00<?, ?B/s]
model.bin: 14%| | 10.5M/75.5M [00:23<02:23, 453kB/s]
model.bin: 14%| | 10.5M/75.5M [00:37<02:23, 453kB/s]
model.bin: 28%| | 21.0M/75.5M [00:53<02:21, 386kB/s]
model.bin: 28%| | 21.0M/75.5M [01:07<02:21, 386kB/s]
model.bin: 42%| | 31.5M/75.5M [01:36<02:24, 304kB/s]
model.bin: 42%| | 31.5M/75.5M [01:47<02:24, 304kB/s]
model.bin: 56%| | 41.9M/75.5M [02:11<01:50, 303kB/s]
model.bin: 56%| | 41.9M/75.5M [02:27<01:50, 303kB/s]
model.bin: 69%| | 52.4M/75.5M [02:34<01:07, 341kB/s]
model.bin: 69%| | 52.4M/75.5M [02:47<01:07, 341kB/s]
model.bin: 83%| | 62.9M/75.5M [02:52<00:31, 400kB/s]
model.bin: 83%| | 62.9M/75.5M [03:07<00:31, 400kB/s]
model.bin: 97%|| 73.4M/75.5M [03:07<00:04, 461kB/s]
model.bin: 97%|| 73.4M/75.5M [03:10<00:05, 386kB/s]
Traceback (most recent call last):
2026-03-04 21:42:57 - faster_whisper - ERROR - File "urllib3\response.py", line 748, in _error_catcher
File "urllib3\response.py", line 894, in _raw_read
urllib3.exceptions.IncompleteRead: IncompleteRead(74480179 bytes read, 1058091 more expected)
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "requests\models.py", line 820, in generate
File "urllib3\response.py", line 1060, in stream
File "urllib3\response.py", line 977, in read
File "urllib3\response.py", line 872, in raw_read
File "contextlib.py", line 153, in exit
2026-03-04 21:42:57 - faster_whisper - ERROR - File "urllib3\response.py", line 772, in error_catcher
urllib3.exceptions.ProtocolError: ('Connection broken: IncompleteRead(74480179 bytes read, 1058091 more expected)', IncompleteRead(74480179 bytes read, 1058091 more expected))
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "D:\whisper-fast-XXL_main.py", line 2245, in
File "D:\whisper-fast-XXL_main.py", line 1598, in cli
File "faster_whisper\utils.py", line 233, in download_model
File "huggingface_hub\utils_validators.py", line 118, in _inner_fn
File "huggingface_hub_snapshot_download.py", line 238, in snapshot_download
File "tqdm\contrib\concurrent.py", line 69, in thread_map
File "tqdm\contrib\concurrent.py", line 51, in _executor_map
File "tqdm\std.py", line 1169, in iter
File "concurrent\futures_base.py", line 621, in result_iterator
File "concurrent\futures_base.py", line 319, in _result_or_cancel
File "concurrent\futures_base.py", line 458, in result
File "concurrent\futures_base.py", line 403, in __get_result
File "concurrent\futures\thread.py", line 58, in run
File "huggingface_hub_snapshot_download.py", line 213, in _inner_hf_hub_download
File "huggingface_hub\utils_validators.py", line 118, in _inner_fn
File "huggingface_hub\file_download.py", line 1461, in hf_hub_download
File "huggingface_hub\file_download.py", line 541, in http_get
File "requests\models.py", line 822, in generate
requests.exceptions.ChunkedEncodingError: ('Connection broken: IncompleteRead(74480179 bytes read, 1058091 more expected)', IncompleteRead(74480179 bytes read, 1058091 more expected))
[PYI-8984:ERROR] Failed to execute script 'main' due to unhandled exception!
Faster Whisper 返回值: 1
2026-03-04 21:42:57 - faster_whisper - ERROR - Faster Whisper 错误: File "urllib3\response.py", line 748, in _error_catcherFile "urllib3\response.py", line 772, in _error_catcher
2026-03-04 21:42:57 - transcript_thread - ERROR - 转录过程中发生错误: File "urllib3\response.py", line 748, in _error_catcherFile "urllib3\response.py", line 772, in _error_catcher
Traceback (most recent call last):
File "D:\Program Files\VideoCaptioner\app\thread\transcript_thread.py", line 101, in run
asr_data = transcribe(
^^^^^^^^^^^
File "D:\Program Files\VideoCaptioner\app\core\bk_asr\transcribe.py", line 86, in transcribe
asr_data = asr.run(callback=callback)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Program Files\VideoCaptioner\app\core\bk_asr\base.py", line 54, in run
resp_data = self._run(callback, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Program Files\VideoCaptioner\app\core\bk_asr\faster_whisper.py", line 266, in _run
raise RuntimeError(error_msg)
RuntimeError: File "urllib3\response.py", line 748, in _error_catcherFile "urllib3\response.py", line 772, in _error_catcher
Latest version info: v1.3.3

[eytion]

补充一下，多次尝试后会出现不同的报错：
开始语音转录
Faster Whisper 执行命令: faster-whisper-xxl -m large-v2 --print_progress --model_dir D:\Program Files\VideoCaptioner\AppData\models C:\Users\zhuyi\AppData\Local\Temp\bk_asr\tmpo9b_qmff\audio.wav -l ja -d cuda --output_format srt -o source --vad_filter true --vad_threshold 0.40 --vad_method silero_v4 --one_word 0 --sentence --max_line_width 30 --max_line_count 1 --max_comma 20 --max_comma_cent 50 --beep_off
Standalone Faster-Whisper-XXL r245.2 running on: CUDA
Starting sequential inference to transcribe: C:\Users\zhuyi\AppData\Local\Temp\bk_asr\tmpo9b_qmff\audio.wav
Traceback (most recent call last):
File "D:\whisper-fast-XXL_main_.py", line 2245, in
File "D:\whisper-fast-XXL_main_.py", line 2118, in cli
File "faster_whisper\transcribe.py", line 2179, in restore_speech_timestamps
File "faster_whisper\transcribe.py", line 1471, in generate_segments
File "faster_whisper\transcribe.py", line 1719, in encode
RuntimeError: cuBLAS failed with status CUBLAS_STATUS_NOT_SUPPORTED
[PYI-6856:ERROR] Failed to execute script 'main' due to unhandled exception!
Faster Whisper 返回值: 1
2026-03-04 21:56:14 - faster_whisper - ERROR - Faster Whisper 错误:
2026-03-04 21:56:14 - transcript_thread - ERROR - 转录过程中发生错误:
Traceback (most recent call last):
File "D:\Program Files\VideoCaptioner\app\thread\transcript_thread.py", line 101, in run
asr_data = transcribe(
^^^^^^^^^^^
File "D:\Program Files\VideoCaptioner\app\core\bk_asr\transcribe.py", line 86, in transcribe
asr_data = asr.run(callback=callback)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Program Files\VideoCaptioner\app\core\bk_asr\base.py", line 54, in run
resp_data = self._run(callback, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Program Files\VideoCaptioner\app\core\bk_asr\faster_whisper.py", line 266, in _run
raise RuntimeError(error_msg)
RuntimeError

[lex0301]

#814 #900