Python利用Pydub实现自动分割音频

pip install pydub

from pydub import AudioSegment
from pydub.silence import split_on_silence

# 读取音频文件
audio = AudioSegment.from_file("audio.mp3", format="mp3")

# 设置分割参数
min_silence_len = 700  # 最小静音长度
silence_thresh =-10  # 静音阈值，越小越严格
keep_silence = 600  # 保留静音长度

# 识别计算分割歌曲数量
num_segments = int(audio.duration_seconds/60/3)  # 每首歌曲大概三分钟，计算歌曲数量

# 分割音频文件
for i in range(-10, 0):
    segments = split_on_silence(audio, min_silence_len=min_silence_len, silence_thresh=i, keep_silence=keep_silence)
    if len(segments) <= num_segments:
        print(f"分割成功，共分割出 {len(segments)} 段")
        break
    else:
        print(f"当前阈值为 {i}，分割出 {len(segments)} 段，继续尝试")

from pydub import AudioSegment
from pydub.utils import make_chunks
import os, re

# # 循环目录下所有文件
for each in os.listdir("D:/PycharmProjects/拾音器/"):  # 循环目录

    filename = re.findall(r"(.*?)\.mp3", each)  # 取出.mp3后缀的文件名
    print(each)
    if each:

        mp3 = AudioSegment.from_file('D:/PycharmProjects/拾音器//{}'.format(each), "mp3")  # 打开mp3文件
        #         # # mp3[17*1000+500:].export(filename[0], format="mp3") # 切割前17.5秒并覆盖保存，与以下代码不可同时使用
        size = 10000  # 切割的毫秒数 10s=10000

        chunks = make_chunks(mp3, size)  # 将文件切割为10s一块

        for i, chunk in enumerate(chunks):
            chunk_name = "{}-{}.mp3".format(each.split(".")[0], i)    # 也可以自定义名字
            print(chunk_name)
            chunk.export('D:/PycharmProjects/拾音器2/{}'.format(chunk_name), format="mp3")  # 新建的保存文件夹

"""
把原音频，按csv中的标注结果分割成几个音频，如text是无，忽略。否则从sDate到eDate 进行分割。
"""
import os
import pandas as pd
import json
from pydub import AudioSegment
 
csv_url = "D:\csv\\"
wav_url = "D:\标注音频与示例\\"
save_wav = "D:\wav\\new_wav\\"
 
for path in os.listdir(csv_url):
    data_frame = pd.read_csv(csv_url + path, encoding='utf-8')
    name_list = data_frame["storeFileName"]
    result_list = data_frame["finalResult"]
    for nl, rl in zip(name_list, result_list):
        new_wav_url = save_wav + nl.split(".")[0] + "\\"
        if not os.path.exists(new_wav_url):
            os.makedirs(new_wav_url)
        result = json.loads(rl)
        txt_url = new_wav_url + nl.split(".")[0]+".txt"
        with open(txt_url, "w", encoding='utf-8') as fn:
            fn.write(nl+"\n")
            fn.write(rl)
            fn.close()
        audio = AudioSegment.from_wav(wav_url + nl)
        for text in result['text']:
            if text['defData']['text'] != '无':
                sTime = text['defData']['sTime'] * 1000
                dTime = text['defData']['dTime'] * 1000
                eTime = sTime + dTime
                # 音频切割按开始时间到结束时间切割
                audio_chunk = audio[sTime:eTime]

                audio_chunk.export(new_wav_url + nl.split(".")[0] + "-" + str(text['id']) + ".wav", format="wav")