关于语音转录使用本地模型时遇到的下载问题

[hisuperyin]

我今天刚刚了解到VideoCaptioner这款很有趣的项目，并在api.videocaptioner.cn充值并填入了api key。
而我目前的需求只需要生成字幕，所以语音转录环节就可以完成。
我目前遇到的问题是公益的语音转录次数已经用完，等我意识到这个问题时，发现只能换成本地（一开始我以为api key也可以用于转录）。
可我使用FasterWhisper接口，然后通过软件自带的下载按钮，无论是下载FasterWhisper程序还是模型，都失败了。
请问这是什么原因？如果没法通过VideoCaptioner自带的下载器，我又应该怎么做呢

[philpw99]

那就需要看看你的程序日志了，不然无法知道在哪里出错。

当然，你也可以自己去直接下载245.2的程序包，
https://github.com/Purfview/whisper-standalone-win/releases
然后解压在 app\resource\bin里面
解压后你会看到这文件夹里面多了个Faster-Whisper-XXL的目录
一旦有了这个目录，VideoCaptioner就会认为faster whisper已经下载了。

[hisuperyin]

感谢你的回复。我使用了直接下载程序包的方案。转录进度没有问题，但最终显示转录失败
我打开“查看日志”，因为本身选的是批量处理，所以只截取了最后一个文件的失败日志
看起来貌似和刚下的faster-whisper-xxl关系不大
日志如下：
===========转录任务开始=========== 时间：2025-04-10 21:14:31.130064 开始转换音频 转换为音频执行命令: ffmpeg -i E:\DATA_Study\考试类\法考\民法公开-李建伟\【视频】2022众合客观题讲座精讲卷-民法李建伟\63【视频】2022众合客观题讲座精讲卷-民法李建伟【一手更新微信s ikao108】.mp4 -map 0:a -ac 1 -ar 16000 -af aresample=async=1 -y C:\Users\YXY\AppData\Local\Temp\tmpd1drxqw7.wav 开始语音转录 Faster Whisper 执行命令: faster-whisper-xxl -m tiny --print_progress --model_dir D:\Software\SmallTool\VideoCaptioner\VideoCaptioner\AppData\models C:\Users\YXY\AppData\Local\Temp\bk_asr\tmp2yzqzjrx\audio.wav -l en -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 90 --max_line_count 1 --max_comma 20 --max_comma_cent 50 --beep_off Model not found at: D:\Software\SmallTool\VideoCaptioner\VideoCaptioner\AppData\models\faster-whisper-tiny Attempting to download: 2025-04-10 21:15:09 - faster_whisper - ERROR - An error occured while synchronizing the model Systran/faster-whisper-tiny from the Hugging Face Hub: (MaxRetryError("HTTPSConnectionPool(host='huggingface.co', port=443): Max retries exceeded with url: /api/models/Systran/faster-whisper-tiny/revision/main (Caused by ConnectTimeoutError(<urllib3.connection.HTTPSConnection object at 0x0000021C9075EB60>, 'Connection to huggingface.co timed out. (connect timeout=None)'))"), '(Request ID: 9cc6a3e4-f371-4ae1-8861-b5a6c7e7c900)') Trying to load the model directly from the local cache, if it exists. Traceback (most recent call last): File "urllib3\connection.py", line 199, in _new_conn File "urllib3\util\connection.py", line 85, in create_connection File "urllib3\util\connection.py", line 73, in create_connection TimeoutError: [WinError 10060] ӷһʱûȷ𸴻ӵûзӦӳʧܡ The above exception was the direct cause of the following exception: Traceback (most recent call last): File "urllib3\connectionpool.py", line 789, in urlopen File "urllib3\connectionpool.py", line 490, in _make_request File "urllib3\connectionpool.py", line 466, in _make_request File "urllib3\connectionpool.py", line 1095, in _validate_conn File "urllib3\connection.py", line 693, in connect File "urllib3\connection.py", line 208, in _new_conn urllib3.exceptions.ConnectTimeoutError: (<urllib3.connection.HTTPSConnection object at 0x0000021C9075EB60>, 'Connection to huggingface.co timed out. (connect timeout=None)') The above exception was the direct cause of the following exception: Traceback (most recent call last): File "requests\adapters.py", line 667, in send File "urllib3\connectionpool.py", line 843, in urlopen File "urllib3\util\retry.py", line 519, in increment urllib3.exceptions.MaxRetryError: HTTPSConnectionPool(host='huggingface.co', port=443): Max retries exceeded with url: /api/models/Systran/faster-whisper-tiny/revision/main (Caused by ConnectTimeoutError(<urllib3.connection.HTTPSConnection object at 0x0000021C9075EB60>, 'Connection to huggingface.co timed out. (connect timeout=None)')) During handling of the above exception, another exception occurred: Traceback (most recent call last): 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 187, in snapshot_download File "huggingface_hub\utils\_validators.py", line 118, in _inner_fn File "huggingface_hub\hf_api.py", line 2112, in repo_info File "huggingface_hub\utils\_validators.py", line 118, in _inner_fn File "huggingface_hub\hf_api.py", line 1921, in model_info File "requests\sessions.py", line 602, in get File "requests\sessions.py", line 589, in request File "requests\sessions.py", line 703, in send File "huggingface_hub\utils\_http.py", line 63, in send File "requests\adapters.py", line 688, in send requests.exceptions.ConnectTimeout: (MaxRetryError("HTTPSConnectionPool(host='huggingface.co', port=443): Max retries exceeded with url: /api/models/Systran/faster-whisper-tiny/revision/main (Caused by ConnectTimeoutError(<urllib3.connection.HTTPSConnection object at 0x0000021C9075EB60>, 'Connection to huggingface.co timed out. (connect timeout=None)'))"), '(Request ID: 9cc6a3e4-f371-4ae1-8861-b5a6c7e7c900)') During handling of the above exception, another exception occurred: Traceback (most recent call last): File "D:\whisper-fast-XXL\__main__.py", line 2245, in <module> File "D:\whisper-fast-XXL\__main__.py", line 1598, in cli File "faster_whisper\utils.py", line 249, in download_model File "huggingface_hub\utils\_validators.py", line 118, in _inner_fn File "huggingface_hub\_snapshot_download.py", line 170, in snapshot_download FileNotFoundError: [Errno 2] No such file or directory: 'D:\\Software\\SmallTool\\VideoCaptioner\\VideoCaptioner\\resource\\bin\\Faster-Whisper-XXL\\.cache\\hub\\models--Systran--faster-whisper-tiny\\refs\\main' [PYI-11844:ERROR] Failed to execute script '__main__' due to unhandled exception! Faster Whisper 返回值: 1 2025-04-10 21:15:10 - faster_whisper - ERROR - Faster Whisper 错误: An error occured while synchronizing the model Systran/faster-whisper-tiny from the Hugging Face Hub: 2025-04-10 21:15:10 - transcript_thread - ERROR - 转录过程中发生错误: An error occured while synchronizing the model Systran/faster-whisper-tiny from the Hugging Face Hub: Traceback (most recent call last): File "D:\Software\SmallTool\VideoCaptioner\VideoCaptioner\app\thread\transcript_thread.py", line 101, in run asr_data = transcribe( ^^^^^^^^^^^ File "D:\Software\SmallTool\VideoCaptioner\VideoCaptioner\app\core\bk_asr\transcribe.py", line 86, in transcribe asr_data = asr.run(callback=callback) ^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\Software\SmallTool\VideoCaptioner\VideoCaptioner\app\core\bk_asr\base.py", line 54, in run resp_data = self._run(callback, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\Software\SmallTool\VideoCaptioner\VideoCaptioner\app\core\bk_asr\faster_whisper.py", line 266, in _run raise RuntimeError(error_msg) RuntimeError: An error occured while synchronizing the model Systran/faster-whisper-tiny from the Hugging Face Hub:

[philpw99]

主要问题出在这里：
faster-whisper-xxl -m tiny --print_progress --model_dir D:\Software\SmallTool\VideoCaptioner\VideoCaptioner\AppData\models C:\Users\YXY\AppData\Local\Temp\bk_asr\tmp2yzqzjrx\audio.wav -l en -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 90 --max_line_count 1 --max_comma 20 --max_comma_cent 50 --beep_off

-m tiny 是错误的，但是当第一次使用卡卡字幕下载模型时，就会出现。
这个命令最近这样的出错报得特别多，看来是个很明显的bug。网友说转换一下转录方式，然后转回FasterWhisper去就可以了。
#436