基于Python 的语音重采样函数

因为工作中会经常遇到不同采样率的声音文件的问题,特意写了一下重采样的程序。
原理就是把采样点转换到时间刻度之后再进行插值,经过测试,是没有问题的。

#!/usr/bin/env python
# -*- coding: utf-8 -*-

# @Time    : 17-7-21 下午2:32
# @Author  : Lei.Jinggui
# @Site    : http://blog.csdn.net/lccever
# @File    : Resample.py
# @Software: PyCharm Community Edition
# @contact: [email protected]

import numpy as np

def Resample(input_signal,src_fs,tar_fs):
    '''

    :param input_signal:输入信号
    :param src_fs:输入信号采样率
    :param tar_fs:输出信号采样率
    :return:输出信号
    '''

    dtype = input_signal.dtype
    audio_len = len(input_signal)
    audio_time_max = 1.0*(audio_len-1) / src_fs
    src_time = 1.0 * np.linspace(0,audio_len,audio_len) / src_fs
    tar_time = 1.0 * np.linspace(0,np.int(audio_time_max*tar_fs),np.int(audio_time_max*tar_fs)) / tar_fs
    output_signal = np.interp(tar_time,src_time,input_signal).astype(dtype)

    return output_signal


if __name__ == '__main__':

    import wave
    import pyaudio

    def playSound(audio_data_short, framerate=16000, channels=1):
        preply = pyaudio.PyAudio()
        # 播放声音
        streamreply = preply.open(format=pyaudio.paInt16,
                                  channels=channels,
                                  rate=framerate,
                                  output=True)
        data = audio_data_short.tostring()
        streamreply.write(data)
        streamreply.close()
        preply.terminate()

    wave_file = 'test.wav'
    audio_file = wave.open(wave_file, 'rb')
    audio_data = audio_file.readframes(audio_file.getnframes())
    audio_data_short = np.fromstring(audio_data, np.short)
    src_fs = audio_file.getframerate()
    src_chanels = audio_file.getnchannels()

    if src_chanels > 1:
        audio_data_short = audio_data_short[::src_chanels]

    tar_fs = np.int(src_fs * 0.5)


    playSound(audio_data_short,framerate=src_fs)

    audio_data_short0 = Resample(audio_data_short,src_fs,tar_fs)

    playSound(audio_data_short0,framerate=tar_fs)

猜你喜欢

转载自blog.csdn.net/lccever/article/details/75646420