脚本：批量修改语音后缀(比如flac转wav) + 对语音进行重采样

一、 批量修改后缀

import os
# 目标文件
filePath = 'E:\\Datasets\\16Khz\\train\\LibriSpeech\\train-clean-100'
# 算法
for i,j,k in os.walk(filePath):
        if len(k)!=0:
                for index in range(len(k)):
                        path = i
                        new_k= k[index].replace('.flac', '.wav')
                        os.rename(os.path.join(path,k[index]),os.path.join(path,new_k))
1
2
3
4
5
6
7
8
9
10

二、批量重采样

import os
import librosa
import soundfile as sf

filePath = 'E:\PaperWork\Codec\datasets'
for i,j,k in os.walk(filePath):
        if len(k)!=0:
                for index in range(len(k)):
                        if(k[index].endswith('.wav')):
                                path = i
                                audio, sr = librosa.load(path+'\\'+k[index])
                                # Resample audio to 8 kHz
                                resampled_audio = librosa.resample(audio, orig_sr=sr, target_sr=8000)
                                # args：保存路径，重采样目标，目标采样率
                                sf.write(path+'\\'+k[index],resampled_audio,samplerate=8000)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15