我在运行示例代码时,上传了一批语音文件(27个wav文件,48KHZ采样率),在预处理步骤,出现报错”ValueError: expect 29 but scp only have 28“, 我检查了一下,这个报错是在执行代码
import os
input_wav = './workspace/Data/Speaker0' # wav audio path work_dir = './workspace/Data/Speaker0_Autolabel' # output path os.makedirs(work_dir, exist_ok=True)
ret, report = run_auto_label(input_wav = input_wav, work_dir = work_dir, resource_revision='v1.0.4') print(report) 时,在make_fbank.py文件中报出来的错误,具体报错代码为: File ~/miniconda3/lib/python3.8/site-packages/tts_autolabel/make_fbank.py:40, in read_scp(wav_scp, expect_num) 38 scp.append(info) 39 if expect_num != len(scp): ---> 40 raise ValueError(f'expect {expect_num} but scp only have {len(scp)}') 41 return scp
请问这是什么原因呢?