通过预加重，分帧，加窗，快速傅里叶变换，求谱线能量，通过梅尔滤波器组得到肺音lung.wav的梅尔谱图，横坐标为lung.wav的时间，纵坐标为频率，python代码

以下是基于Python的肺音梅尔谱图提取代码： ```python import numpy as np import scipy.signal as signal import scipy.io.wavfile as wavfile # 定义预处理参数 pre_emphasis = 0.97 frame_size = 0.025 frame_stride = 0.01 NFFT = 512 nfilt = 40 # 定义梅尔滤波器组参数 low_freq_mel = 0 high_freq_mel = (2595 * np.log10(1 + (wavfile.read('lung.wav')[0] / 2) / 700)) # 根据Nyquist频率计算高频 mel_points = np.linspace(low_freq_mel, high_freq_mel, nfilt + 2) # 均匀分布梅尔刻度 hz_points = (700 * (10 ** (mel_points / 2595) - 1)) # 将梅尔刻度转换为赫兹刻度 bin = np.floor((NFFT + 1) * hz_points / wavfile.read('lung.wav')[0]) # 定义加窗函数 hamming = np.hamming # 加载音频文件 signal, sr = wavfile.read('lung.wav') # 预加重 emphasized_signal = np.append(signal[0], signal[1:] - pre_emphasis * signal[:-1]) # 分帧 frame_length, frame_step = frame_size * sr, frame_stride * sr signal_length = len(emphasized_signal) frame_length = int(round(frame_length)) frame_step = int(round(frame_step)) num_frames = int(np.ceil(float(np.abs(signal_length - frame_length)) / frame_step)) pad_signal_length = num_frames * frame_step + frame_length z = np.zeros((pad_signal_length - signal_length)) pad_signal = np.append(emphasized_signal, z) indices = np.tile(np.arange(0, frame_length), (num_frames, 1)) + np.tile( np.arange(0, num_frames * frame_step, frame_step), (frame_length, 1)).T frames = pad_signal[indices.astype(np.int32, copy=False)] # 加窗 frames *= hamming(frame_length) # 快速傅里叶变换 mag_frames = np.absolute(np.fft.rfft(frames, NFFT)) # Magnitude of the FFT pow_frames = ((1.0 / NFFT) * ((mag_frames) ** 2)) # Power Spectrum # 建立梅尔滤波器组 f_bank = np.zeros((nfilt, int(np.floor(NFFT / 2 + 1)))) for m in range(1, nfilt + 1): f_m_minus = int(bin[m - 1]) # left f_m = int(bin[m]) # center f_m_plus = int(bin[m + 1]) # right for k in range(f_m_minus, f_m): f_bank[m - 1, k] = (k - bin[m - 1]) / (bin[m] - bin[m - 1]) for k in range(f_m, f_m_plus): f_bank[m - 1, k] = (bin[m + 1] - k) / (bin[m + 1] - bin[m]) # 应用梅尔滤波器组 filter_banks = np.dot(pow_frames, f_bank.T) # 取对数 filter_banks = np.where(filter_banks == 0, np.finfo(float).eps, filter_banks) # 防止出现log(0)的情况 filter_banks = 20 * np.log10(filter_banks) # dB # 可视化梅尔谱图 import matplotlib.pyplot as plt plt.imshow(filter_banks.T, cmap="hot", origin="lower") plt.xlabel("Time") plt.ylabel("Frequency") plt.title("Mel Spectrogram") plt.colorbar() plt.show() ``` 运行以上代码后将会生成肺音lung.wav的梅尔谱图。