audio_conv_utils.py 文件源码

def preprocess_input(audio_path, dim_ordering='default'):
    '''Reads an audio file and outputs a Mel-spectrogram.
    '''
    if dim_ordering == 'default':
        dim_ordering = K.image_dim_ordering()
    assert dim_ordering in {'tf', 'th'}

    if librosa_exists():
        import librosa
    else:
        raise RuntimeError('Librosa is required to process audio files.\n' +
                           'Install it via `pip install librosa` \nor visit ' +
                           'http://librosa.github.io/librosa/ for details.')

    # mel-spectrogram parameters
    SR = 12000
    N_FFT = 512
    N_MELS = 96
    HOP_LEN = 256
    DURA = 29.12

    src, sr = librosa.load(audio_path, sr=SR)
    n_sample = src.shape[0]
    n_sample_wanted = int(DURA * SR)

    # trim the signal at the center
    if n_sample < n_sample_wanted:  # if too short
        src = np.hstack((src, np.zeros((int(DURA * SR) - n_sample,))))
    elif n_sample > n_sample_wanted:  # if too long
        src = src[(n_sample - n_sample_wanted) / 2:
                  (n_sample + n_sample_wanted) / 2]

    logam = librosa.logamplitude
    melgram = librosa.feature.melspectrogram
    x = logam(melgram(y=src, sr=SR, hop_length=HOP_LEN,
                      n_fft=N_FFT, n_mels=N_MELS) ** 2,
              ref_power=1.0)

    if dim_ordering == 'th':
        x = np.expand_dims(x, axis=0)
    elif dim_ordering == 'tf':
        x = np.expand_dims(x, axis=3)
    return x