音频数据预处理和序列化脚本

这是一个 Python 脚本，用于将音频数据预处理成适合机器学习的格式。它将音频文件切分成片段，并将其序列化为 NumPy 数组，以便于模型训练。

import os
import librosa
import numpy as np
from tqdm import tqdm

clean_train_folder = 'data/clean_trainset_56spk_wav/clean_trainset_56spk_wav'
noisy_train_folder = 'data/noisy_trainset_56spk_wav/noisy_trainset_56spk_wav'
clean_test_folder = 'data/clean_testset_wav/clean_testset_wav'
noisy_test_folder = 'data/noisy_testset_wav/noisy_testset_wav'
serialized_train_folder = 'data/serialized_train_data'
serialized_test_folder = 'data/serialized_test_data'
window_size = 2 ** 14  # about 1 second of samples
sample_rate = 16000


def slice_signal(file, window_size, stride, sample_rate):
    '''
    Helper function for slicing the audio file
    by window size and sample rate with [1-stride] percent overlap (default 50%).
    '''
    wav, sr = librosa.load(file, sr=sample_rate)
    hop = int(window_size * stride)
    slices = []
    for end_idx in range(window_size, len(wav), hop):
        start_idx = end_idx - window_size
        slice_sig = wav[start_idx:end_idx]
        slices.append(slice_sig)
    return slices


def process_and_serialize(data_type):
    '''
    Serialize, down-sample the sliced signals and save on separate folder.
    '''
    stride = 0.5

    if data_type == 'train':
        clean_folder = clean_train_folder
        noisy_folder = noisy_train_folder
        serialized_folder = serialized_train_folder
    else:
        clean_folder = clean_test_folder
        noisy_folder = noisy_test_folder
        serialized_folder = serialized_test_folder
    if not os.path.exists(serialized_folder):
        os.makedirs(serialized_folder)

    # walk through the path, slice the audio file, and save the serialized result
    for root, dirs, files in os.walk(clean_folder):
        if len(files) == 0:
            continue
        for filename in tqdm(files, desc='Serialize and down-sample {} audios'.format(data_type)):
            clean_file = os.path.join(clean_folder, filename)
            noisy_file = os.path.join(noisy_folder, filename)
            # slice both clean signal and noisy signal
            clean_sliced = slice_signal(clean_file, window_size, stride, sample_rate)
            noisy_sliced = slice_signal(noisy_file, window_size, stride, sample_rate)
            # serialize - file format goes [original_file]_[slice_number].npy
            # ex) p293_154.wav_5.npy denotes 5th slice of p293_154.wav file
            for idx, slice_tuple in enumerate(zip(clean_sliced, noisy_sliced)):
                pair = np.array([slice_tuple[0], slice_tuple[1]])
                np.save(os.path.join(serialized_folder, '{}_{}'.format(filename, idx)), arr=pair)


def data_verify(data_type):
    '''
    Verifies the length of each data after pre-process.
    '''
    if data_type == 'train':
        serialized_folder = serialized_train_folder
    else:
        serialized_folder = serialized_test_folder

    for root, dirs, files in os.walk(serialized_folder):
        for filename in tqdm(files, desc='Verify serialized {} audios'.format(data_type)):
            data_pair = np.load(os.path.join(root, filename))
            if data_pair.shape[1] != window_size:
                print('Snippet length not {} : {} instead'.format(window_size, data_pair.shape[1]))
                break


if __name__ == '__main__':
    process_and_serialize('train')
    data_verify('train')
    process_and_serialize('test')
    data_verify('test')

不建议删除生成的文件的一部分，因为这可能会导致数据不完整或不正确。如果需要减少数据量，可以在初始文件夹中删除一些音频文件，然后重新运行脚本来生成新的序列化文件。