Python pydub音频处理

李小白是一只喵

发布于 2020-04-23 15:02:22

2.2K0

发布于 2020-04-23 15:02:22

文章被收录于专栏：算法微时光算法微时光

image.png

pydub

Pydub可以让你用简单的方式处理音频。

Pydub提供了简洁的高层接口，极大的扩展了python处理音频文件的能力。

GitHub链接：pydub-github

GitHub:

You can open and save WAV files with pure python. For opening and saving non-wav files – like mp3 – you'll need ffmpeg or libav.

安装

pip install pydub

image.png

安装依赖软件

pydub的使用必须安装对应的依赖软件ffmpeg或avconv

# ffmpeg
sudo apt-get install ffmpeg libavcodec-extra

验证是否安装成功：

image.png

API函数使用

打开接口

Open a WAV file

from pydub import AudioSegment

song = AudioSegment.from_wav("never_gonna_give_you_up.wav")

Open a mp3 file

song = AudioSegment.from_mp3("never_gonna_give_you_up.mp3")

Open a other file

ogg_version = AudioSegment.from_ogg("never_gonna_give_you_up.ogg")
flv_version = AudioSegment.from_flv("never_gonna_give_you_up.flv")

mp4_version = AudioSegment.from_file("never_gonna_give_you_up.mp4", "mp4")
wma_version = AudioSegment.from_file("never_gonna_give_you_up.wma", "wma")
aac_version = AudioSegment.from_file("never_gonna_give_you_up.aiff", "aac")

音频操作

切割音频

# pydub does things in milliseconds
ten_seconds = 10 * 1000

first_10_seconds = song[:ten_seconds]

last_5_seconds = song[-5000:]

分贝操作 分贝（decibel）是量度两个相同单位之数量比例的计量单位，主要用于度量声音强度，常用dB表示。

# boost volume by 6dB
beginning = first_10_seconds + 6

# reduce volume by 3dB
end = last_5_seconds - 3

分贝	说明
1分贝	刚能听到的声音
15 分贝以下	感觉安静
30 分贝	耳语的音量大小
40 分贝	冰箱的嗡嗡声
60分贝	正常交谈的声音
70分贝	相当于走在闹市区
85分贝	汽车穿梭的马路上
95分贝	摩托车启动声音
100分贝	装修电钻的声音
110分贝	卡拉OK、大声播放MP3 的声音
120分贝	飞机起飞时的声音
150分贝	燃放烟花爆竹的声音

音频链接 将一个文件添加到另一个文件的末尾

without_the_middle = beginning + end

音频长度

without_the_middle.duration_seconds == 15.0

淡入淡出

# 1.5 second crossfade
with_style = beginning.append(end, crossfade=1500)

重复音频

# repeat the clip twice
do_it_over = with_style * 2

再次淡入淡出

# 2 sec fade in, 3 sec fade out
awesome = do_it_over.fade_in(2000).fade_out(3000)

保存音频

直接保存 所有ffmpeg支持的都支持

awesome.export("mashup.mp3", format="mp3")

用标签保存结果（元数据）

awesome.export("mashup.mp3", format="mp3", tags={'artist': 'Various artists', 'album': 'Best of 2011', 'comments': 'This album is awesome!'})

实例：

# -*- coding: utf-8 -*-
from glob import glob
from pydub import AudioSegment

playlist_songs = [AudioSegment.from_mp3(mp3_file) for mp3_file in glob("*.mp3")]

first_song = playlist_songs.pop(0)

# let's just include the first 30 seconds of the first song (slicing
# is done by milliseconds)
beginning_of_song = first_song[:30*1000]

playlist = beginning_of_song
for song in playlist_songs:

    # We don't want an abrupt stop at the end, so let's do a 10 second crossfades
    playlist = playlist.append(song, crossfade=(10 * 1000))

# let's fade out the end of the last song
playlist = playlist.fade_out(30)

# hmm I wonder how long it is... ( len(audio_segment) returns milliseconds )
playlist_length = len(playlist) / (1000*60)

# lets save it!
out_f = open("%s_minute_playlist.mp3" % playlist_length, 'wb')

playlist.export(out_f, format='mp3')

实战

将mp3文件转换成wav文件：

# -*- coding: utf-8 -*-
from pydub import AudioSegment
 
def trans_mp3_to_wav(filepath):
    song = AudioSegment.from_mp3(filepath)
    song.export("out.wav", format="wav")

if __name__ == "__main__":
    trans_mp3_to_wav("一剪梅.mp3")