data_utils/wav2mel.py

import librosa
import librosa.filters
import numpy as np
from scipy import signal
from wav2mel_hparams import hparams as hp
from librosa.core.audio import resample
import soundfile as sf

def load_wav(path, sr):
    return librosa.core.load(path, sr=sr)

def preemphasis(wav, k, preemphasize=True):
    if preemphasize:
        return signal.lfilter([1, -k], [1], wav)
    return wav

def inv_preemphasis(wav, k, inv_preemphasize=True):
    if inv_preemphasize:
        return signal.lfilter([1], [1, -k], wav)
    return wav

def get_hop_size():
    hop_size = hp.hop_size
    if hop_size is None:
        assert hp.frame_shift_ms is not None
        hop_size = int(hp.frame_shift_ms / 1000 * hp.sample_rate)
    return hop_size

def linearspectrogram(wav):
    D = _stft(preemphasis(wav, hp.preemphasis, hp.preemphasize))
    S = _amp_to_db(np.abs(D)) - hp.ref_level_db
    
    if hp.signal_normalization:
        return _normalize(S)
    return S

def melspectrogram(wav):
    D = _stft(preemphasis(wav, hp.preemphasis, hp.preemphasize))
    S = _amp_to_db(_linear_to_mel(np.abs(D))) - hp.ref_level_db
    
    if hp.signal_normalization:
        return _normalize(S)
    return S

def _stft(y):
    return librosa.stft(y=y, n_fft=hp.n_fft, hop_length=get_hop_size(), win_length=hp.win_size)

##########################################################
#Those are only correct when using lws!!! (This was messing with Wavenet quality for a long time!)
def num_frames(length, fsize, fshift):
    """Compute number of time frames of spectrogram
    """
    pad = (fsize - fshift)
    if length % fshift == 0:
        M = (length + pad * 2 - fsize) // fshift + 1
    else:
        M = (length + pad * 2 - fsize) // fshift + 2
    return M


def pad_lr(x, fsize, fshift):
    """Compute left and right padding
    """
    M = num_frames(len(x), fsize, fshift)
    pad = (fsize - fshift)
    T = len(x) + 2 * pad
    r = (M - 1) * fshift + fsize - T
    return pad, pad + r
##########################################################
#Librosa correct padding
def librosa_pad_lr(x, fsize, fshift):
    return 0, (x.shape[0] // fshift + 1) * fshift - x.shape[0]

# Conversions
_mel_basis = None

def _linear_to_mel(spectogram):
    global _mel_basis
    if _mel_basis is None:
        _mel_basis = _build_mel_basis()
    return np.dot(_mel_basis, spectogram)

def _build_mel_basis():
    assert hp.fmax <= hp.sample_rate // 2
    return librosa.filters.mel(sr=hp.sample_rate, n_fft=hp.n_fft, n_mels=hp.num_mels,
                               fmin=hp.fmin, fmax=hp.fmax)

def _amp_to_db(x):
    min_level = np.exp(hp.min_level_db / 20 * np.log(10))
    return 20 * np.log10(np.maximum(min_level, x))

def _db_to_amp(x):
    return np.power(10.0, (x) * 0.05)

def _normalize(S):
    if hp.allow_clipping_in_normalization:
        if hp.symmetric_mels:
            return np.clip((2 * hp.max_abs_value) * ((S - hp.min_level_db) / (-hp.min_level_db)) - hp.max_abs_value,
                           -hp.max_abs_value, hp.max_abs_value)
        else:
            return np.clip(hp.max_abs_value * ((S - hp.min_level_db) / (-hp.min_level_db)), 0, hp.max_abs_value)
    
    assert S.max() <= 0 and S.min() - hp.min_level_db >= 0
    if hp.symmetric_mels:
        return (2 * hp.max_abs_value) * ((S - hp.min_level_db) / (-hp.min_level_db)) - hp.max_abs_value
    else:
        return hp.max_abs_value * ((S - hp.min_level_db) / (-hp.min_level_db))

def _denormalize(D):
    if hp.allow_clipping_in_normalization:
        if hp.symmetric_mels:
            return (((np.clip(D, -hp.max_abs_value,
                              hp.max_abs_value) + hp.max_abs_value) * -hp.min_level_db / (2 * hp.max_abs_value))
                    + hp.min_level_db)
        else:
            return ((np.clip(D, 0, hp.max_abs_value) * -hp.min_level_db / hp.max_abs_value) + hp.min_level_db)
    
    if hp.symmetric_mels:
        return (((D + hp.max_abs_value) * -hp.min_level_db / (2 * hp.max_abs_value)) + hp.min_level_db)
    else:
        return ((D * -hp.min_level_db / hp.max_abs_value) + hp.min_level_db)



def wav2mel(wav, sr):
        wav16k = resample(wav, orig_sr=sr, target_sr=16000)
        # print('wav16k', wav16k.shape, wav16k.dtype)
        mel = melspectrogram(wav16k)
        # print('mel', mel.shape, mel.dtype)
        if np.isnan(mel.reshape(-1)).sum() > 0:
            raise ValueError(
                'Mel contains nan! Using a TTS voice? Add a small epsilon noise to the wav file and try again')
        # mel.dtype = np.float32
        mel_chunks = []
        mel_idx_multiplier = 80. / 25
        mel_step_size = 8
        i = start_idx = 0
        while start_idx < len(mel[0]):
            start_idx = int(i * mel_idx_multiplier)
            if start_idx + mel_step_size // 2 > len(mel[0]):
                mel_chunks.append(mel[:, len(mel[0]) - mel_step_size:])
            elif start_idx - mel_step_size // 2 < 0:
                mel_chunks.append(mel[:, :mel_step_size])
            else:
                mel_chunks.append(mel[:, start_idx - mel_step_size // 2 : start_idx + mel_step_size // 2])
            i += 1
        return mel_chunks



if __name__ == '__main__':
    import argparse

    parser = argparse.ArgumentParser()
    parser.add_argument('--wav', type=str, default='')
    parser.add_argument('--save_feats', action='store_true')

    opt = parser.parse_args()

    wav, sr = librosa.core.load(opt.wav)
    mel_chunks = np.array(wav2mel(wav.T, sr))
    print(mel_chunks.shape, mel_chunks.transpose(0,2,1).shape)
    
    if opt.save_feats:
        save_path = opt.wav.replace('.wav', '_mel.npy')
        np.save(save_path, mel_chunks.transpose(0,2,1))
        print(f"[INFO] saved logits to {save_path}")