1
                """
                Python绘制语谱图
                """
                 2
                """
                Python绘制时域波形
                """
                 3
                 4
                #
                 导入相应的包
                 5
                import
                 numpy, wave

                 6
                import
                 matplotlib.pyplot as plt

                 7
                import
                 numpy as np

                 8
                import
                 os

                 9
                10 filepath = ‘G:/实战培训/Python生成语谱图/ReNoise/Prim10/‘# 添加路径11 filename = os.listdir(filepath)  # 得到文件夹下的所有文件名1213for i in range(len(filename)):
14     f = wave.open(filepath + filename[i], ‘rb‘)  # 调用wave模块中的open函数，打开语音文件。15     params = f.getparams()  # 得到语音参数16     nchannels, sampwidth, framerate, nframes = params[:4]  # nchannels:音频通道数，sampwidth:每个音频样本的字节数，framerate:采样率，nframes:音频采样点数17     strData = f.readframes(nframes)  # 读取音频，字符串格式18     wavaData = np.fromstring(strData, dtype=np.int16)  # 得到的数据是字符串，将字符串转为int型19     wavaData = wavaData * 1.0/max(abs(wavaData))  # wave幅值归一化20     wavaData = np.reshape(wavaData, [nframes, nchannels]).T  # .T 表示转置21    f.close()
2223#（1）绘制语谱图24    plt.figure()
25     plt.specgram(wavaData[0], Fs=framerate, scale_by_freq=True, sides=‘default‘)  # 绘制频谱26     plt.xlabel(‘Time(s)‘)
27     plt.ylabel(‘Frequency‘)
28     plt.title("Spectrogram_{}".format(i+1))
29     plt.savefig(‘G:/实战培训/Python生成语谱图/语谱图/{}.jpg‘.format(filename[i][:-4]))
30    plt.show()
3132#（2）绘制时域波形33     time = np.arange(0, nframes) * (1.0 / framerate)
34     time = np.reshape(time, [nframes, 1]).T
35     plt.plot(time[0, :nframes], wavaData[0, :nframes], c="b")
36     plt.xlabel("time(seconds)")
37     plt.ylabel("amplitude")
38     plt.title("Original wave")
39     plt.savefig(‘G:/实战培训/Python生成语谱图/语谱图/{}_.jpg‘.format(filename[i][:-4]))  # 保存绘制的图形40     plt.show()