Python3分析处理声音数据的实例讲解

#!/usr/bin/env python
# encoding: utf-8
"""
@Company：华中科技大学电气学院聚变与等离子研究所
@version: V1.0
@author: Victor
@contact: 1650996069@qq.com or yexin@hust.edu.cn 2018--2020
@software: PyCharm
@file: VoiceDataProcessing.py
@time: 2018/9/8 13:08
@Desc：处理声音数据
"""
 
import wave
import pylab as pl
import numpy as np
 
print('1650996069@qq.com or yexin@hust.edu.cn')
 
print('working...')
 
print("read wav data....")
# 打开WAV文档,声音文件
f = wave.open(r"speak.wav", "rb")
fo = wave.open(r"jg.wav", "wb")
fi = wave.open(r"back.wav", "rb")
 
# 读取波形数据
# (nchannels, sampwidth, framerate, nframes, comptype, compname)
params = f.getparams()
nchannels, sampwidth, framerate, nframes = params[:4]
str_data = f.readframes(nframes)
 
fi_params = fi.getparams()
fi_nframes = fi_params[3]
fi_str_data = fi.readframes(fi_nframes)
 
# 将波形数据转换为数组，并更改
print("update wav data....")
wave_data = np.fromstring(str_data, dtype=np.short)
fi_wave_data = np.fromstring(fi_str_data, dtype=np.short)
 
# 复制并合并声音
new_wave_data = np.hstack((wave_data, wave_data, wave_data, wave_data, wave_data, wave_data, wave_data, wave_data))
temp_wavedata = fi_wave_data[:len(new_wave_data)]
new_wave_data = temp_wavedata * 0.5 + new_wave_data * 0.5
 
new_wave_data = np.array(new_wave_data)
new_wave_data = new_wave_data.astype(wave_data.dtype)
new_str_data = new_wave_data.tostring()
# 写波形数据参数
print("save new wav files....")
fo.setnchannels(nchannels)
fo.setframerate(framerate)
fo.setsampwidth(sampwidth)
fo.writeframes(new_str_data)
 
# 绘制波形
wave_data.shape = -1, 2
wave_data = wave_data.T
time = np.arange(0, nframes) * (1.0 / framerate)
pl.subplot(221)
pl.plot(time, wave_data[0])
pl.subplot(222)
pl.plot(time, wave_data[1], c="g")
pl.xlabel("time (seconds)")
pl.show()
 
# 绘制波形
new_wave_data.shape = -1, 2
new_wave_data = new_wave_data.T
new_time = np.arange(0, nframes * 8) * (1.0 / framerate)
pl.subplot(223)
pl.plot(new_time, new_wave_data[0])
pl.subplot(224)
pl.plot(new_time, new_wave_data[1], c="g")
pl.xlabel("time (seconds)")
pl.show()