源码运行Numpy报错

[Rakshesha2024]

WhisperX model loading error:Numpy is not available Make sure you have downloaded the model first. WhisperX processing error: Numpy is not available 2024-10-02 15:36:58.586 Uncaught app exception Traceback (most recent call last): File "D:\anaconda3\envs\videolingo\lib\site-packages\streamlit\runtime\scriptrunner\exec_code.py", line 88, in exec_func_with_error_handling result = func() File "D:\anaconda3\envs\videolingo\lib\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 579, in code_to_exec exec(code, module.dict) File "D:\AIGC\VideoLingo\st.py", line 125, in main() File "D:\AIGC\VideoLingo\st.py", line 121, in main text_processing_section() File "D:\AIGC\VideoLingo\st.py", line 28, in text_processing_section process_text() File "D:\AIGC\VideoLingo\st.py", line 50, in process_text step2_whisper.transcribe(video_file) File "D:\AIGC\VideoLingo\core\step2_whisper.py", line 20, in transcribe ts(video_file) File "D:\AIGC\VideoLingo\core\all_whisper_methods\whisperX.py", line 124, in transcribe result = transcribe_audio(audio_file, start, end) File "D:\AIGC\VideoLingo\core\all_whisper_methods\whisperX.py", line 45, in transcribe_audio model = whisperx.load_model(model_name, device, compute_type=compute_type, download_root=whisperx_model_dir) File "d:\aigc\videolingo\third_party\whisperx\whisperx\asr.py", line 347, in load_model vad_model = load_vad_model(torch.device(device), use_auth_token=None, *default_vad_options) File "d:\aigc\videolingo\third_party\whisperx\whisperx\vad.py", line 51, in load_vad_model vad_model = Model.from_pretrained(model_fp, use_auth_token=use_auth_token) File "D:\anaconda3\envs\videolingo\lib\site-packages\pyannote\audio\core\model.py", line 720, in from_pretrained raise e File "D:\anaconda3\envs\videolingo\lib\site-packages\pyannote\audio\core\model.py", line 696, in from_pretrained model = Klass.load_from_checkpoint( File "D:\anaconda3\envs\videolingo\lib\site-packages\pytorch_lightning\utilities\model_helpers.py", line 125, in wrapper return self.method(cls, args, kwargs) File "D:\anaconda3\envs\videolingo\lib\site-packages\pytorch_lightning\core\module.py", line 1586, in load_from_checkpoint loaded = _load_from_checkpoint( File "D:\anaconda3\envs\videolingo\lib\site-packages\pytorch_lightning\core\saving.py", line 91, in _load_from_checkpoint model = _load_state(cls, checkpoint, strict=strict, kwargs) File "D:\anaconda3\envs\videolingo\lib\site-packages\pytorch_lightning\core\saving.py", line 165, in _load_state obj = instantiator(cls, _cls_kwargs) if instantiator else cls(_cls_kwargs) File "D:\anaconda3\envs\videolingo\lib\site-packages\pyannote\audio\models\segmentation\PyanNet.py", line 92, in init self.sincnet = SincNet(self.hparams.sincnet) File "D:\anaconda3\envs\videolingo\lib\site-packages\pyannote\audio\models\blocks\sincnet.py", line 52, in init ParamSincFB( File "D:\anaconda3\envs\videolingo\lib\site-packages\asteroid_filterbanks\param_sinc_fb.py", line 56, in init self._initialize_filters() File "D:\anaconda3\envs\videolingo\lib\site-packages\asteroid_filterbanks\param_sinc_fb.py", line 79, in _initialize_filters self.lowhz = nn.Parameter(torch.from_numpy(hz[:-1]).view(-1, 1)) RuntimeError: Numpy is not available

Numpy更换过2.0.0和1.26.4版本，都会跳出这个报错

[Huanshere]

我使用的是1.26.4，我检查一下其他的依赖，谢谢反馈

[SomeoneKong]

2.0似乎是不行的，换成了1.26.4就可以跑了

[Huanshere]

在 43a8142 中已固定版本 谢谢反馈