/content/output_training_data/wav_pre/temp.wav

[JavaAndPython55]

按照你的一步步来的，结果到这里没有这个temp.wav文件提示，报错

[JavaAndPython55]

执行这段代码的时候input_wav = "./test_wavs/" output_data = "./output_training_data/"

ret, report = run_auto_label(input_wav=input_wav, work_dir=output_data, resource_revision="v1.0.7")提示

[JavaAndPython55]

2023-09-06 07:29:24,634 - modelscope - INFO - Use user-specified model revision: v1.0.7 --- Remove /content/output_training_data/paragraph/prosody folder! --- --- New folder /content/output_training_data/paragraph/prosody... --- --- OK --- --- New folder /content/output_training_data/wav... --- --- OK --- --- Remove /content/output_training_data/wav_decrease_sp folder! --- --- New folder /content/output_training_data/wav_decrease_sp... --- --- OK --- --- Remove /content/output_training_data/log folder! --- --- New folder /content/output_training_data/log... --- --- OK --- 2023-09-06 07:29:39 TTS-AutoLabel version: 1.1.8 TTS-AutoLabel resource path: /root/.cache/modelscope/hub/damo/speech_ptts_autolabel_16k/model Target sampling rate: 16000 Input wav dir: /content/test_wavs Output data dir: /content/output_training_data wav_preprocess start... --- There is this folder! --- 0%| | 0/35 [00:00<?, ?it/s]

FileNotFoundError Traceback (most recent call last) in <cell line: 4>() 2 output_data = "./output_training_data/" 3 ----> 4 ret, report = run_auto_label(input_wav=input_wav, work_dir=output_data, resource_revision="v1.0.7")

3 frames /usr/local/lib/python3.10/dist-packages/tts_autolabel/resample_16.py in resample(src_wav_dir, tgt_wav_dir, tgt_sr) 58 logging.info(cmdline) 59 os.system(cmdline) ---> 60 os.remove(temp_file) 61 62

FileNotFoundError: [Errno 2] No such file or directory: '/content/output_training_data/wav_pre/temp.wav'

[JavaAndPython55]

解决了，这个必须要在pytorch的那个目录里面创建文件夹和执行

[venxzw]

@JavaAndPython55 请问是创建那三个文件的时候需要进入到pytorch_wavelets这个文件夹里吗？

[luhairong11]

我也遇到这个问题，工程下没有pytorch的目录呀