init ascend tts

ascend_910-gpt-sovits/GPT-SoVITS/GPT_SoVITS/module/mel_processing.py

@@ -0,0 +1,143 @@
+import torch
+import torch.utils.data
+from librosa.filters import mel as librosa_mel_fn
+
+MAX_WAV_VALUE = 32768.0
+
+
+def dynamic_range_compression_torch(x, C=1, clip_val=1e-5):
+    """
+    PARAMS
+    ------
+    C: compression factor
+    """
+    return torch.log(torch.clamp(x, min=clip_val) * C)
+
+
+def dynamic_range_decompression_torch(x, C=1):
+    """
+    PARAMS
+    ------
+    C: compression factor used to compress
+    """
+    return torch.exp(x) / C
+
+
+def spectral_normalize_torch(magnitudes):
+    output = dynamic_range_compression_torch(magnitudes)
+    return output
+
+
+def spectral_de_normalize_torch(magnitudes):
+    output = dynamic_range_decompression_torch(magnitudes)
+    return output
+
+
+mel_basis = {}
+hann_window = {}
+
+
+def spectrogram_torch(y, n_fft, sampling_rate, hop_size, win_size, center=False):
+    if torch.min(y) < -1.2:
+        print("min value is ", torch.min(y))
+    if torch.max(y) > 1.2:
+        print("max value is ", torch.max(y))
+
+    global hann_window
+    dtype_device = str(y.dtype) + "_" + str(y.device)
+    # wnsize_dtype_device = str(win_size) + '_' + dtype_device
+    key = "%s-%s-%s-%s-%s" % (dtype_device, n_fft, sampling_rate, hop_size, win_size)
+    # if wnsize_dtype_device not in hann_window:
+    if key not in hann_window:
+        # hann_window[wnsize_dtype_device] = torch.hann_window(win_size).to(dtype=y.dtype, device=y.device)
+        hann_window[key] = torch.hann_window(win_size).to(dtype=y.dtype, device=y.device)
+
+    y = torch.nn.functional.pad(
+        y.unsqueeze(1), (int((n_fft - hop_size) / 2), int((n_fft - hop_size) / 2)), mode="reflect"
+    )
+    y = y.squeeze(1)
+    # spec = torch.stft(y, n_fft, hop_length=hop_size, win_length=win_size, window=hann_window[wnsize_dtype_device],
+    spec = torch.stft(
+        y,
+        n_fft,
+        hop_length=hop_size,
+        win_length=win_size,
+        window=hann_window[key],
+        center=center,
+        pad_mode="reflect",
+        normalized=False,
+        onesided=True,
+        return_complex=False,
+    )
+
+    spec = torch.sqrt(spec.pow(2).sum(-1) + 1e-8)
+    return spec
+
+
+def spec_to_mel_torch(spec, n_fft, num_mels, sampling_rate, fmin, fmax):
+    global mel_basis
+    dtype_device = str(spec.dtype) + "_" + str(spec.device)
+    # fmax_dtype_device = str(fmax) + '_' + dtype_device
+    key = "%s-%s-%s-%s-%s-%s" % (dtype_device, n_fft, num_mels, sampling_rate, fmin, fmax)
+    # if fmax_dtype_device not in mel_basis:
+    if key not in mel_basis:
+        mel = librosa_mel_fn(sr=sampling_rate, n_fft=n_fft, n_mels=num_mels, fmin=fmin, fmax=fmax)
+        # mel_basis[fmax_dtype_device] = torch.from_numpy(mel).to(dtype=spec.dtype, device=spec.device)
+        mel_basis[key] = torch.from_numpy(mel).to(dtype=spec.dtype, device=spec.device)
+    # spec = torch.matmul(mel_basis[fmax_dtype_device], spec)
+    spec = torch.matmul(mel_basis[key], spec)
+    spec = spectral_normalize_torch(spec)
+    return spec
+
+
+def mel_spectrogram_torch(y, n_fft, num_mels, sampling_rate, hop_size, win_size, fmin, fmax, center=False):
+    if torch.min(y) < -1.2:
+        print("min value is ", torch.min(y))
+    if torch.max(y) > 1.2:
+        print("max value is ", torch.max(y))
+
+    global mel_basis, hann_window
+    dtype_device = str(y.dtype) + "_" + str(y.device)
+    # fmax_dtype_device = str(fmax) + '_' + dtype_device
+    fmax_dtype_device = "%s-%s-%s-%s-%s-%s-%s-%s" % (
+        dtype_device,
+        n_fft,
+        num_mels,
+        sampling_rate,
+        hop_size,
+        win_size,
+        fmin,
+        fmax,
+    )
+    # wnsize_dtype_device = str(win_size) + '_' + dtype_device
+    wnsize_dtype_device = fmax_dtype_device
+    if fmax_dtype_device not in mel_basis:
+        mel = librosa_mel_fn(sr=sampling_rate, n_fft=n_fft, n_mels=num_mels, fmin=fmin, fmax=fmax)
+        mel_basis[fmax_dtype_device] = torch.from_numpy(mel).to(dtype=y.dtype, device=y.device)
+    if wnsize_dtype_device not in hann_window:
+        hann_window[wnsize_dtype_device] = torch.hann_window(win_size).to(dtype=y.dtype, device=y.device)
+
+    y = torch.nn.functional.pad(
+        y.unsqueeze(1), (int((n_fft - hop_size) / 2), int((n_fft - hop_size) / 2)), mode="reflect"
+    )
+    y = y.squeeze(1)
+
+    spec = torch.stft(
+        y,
+        n_fft,
+        hop_length=hop_size,
+        win_length=win_size,
+        window=hann_window[wnsize_dtype_device],
+        center=center,
+        pad_mode="reflect",
+        normalized=False,
+        onesided=True,
+        return_complex=False,
+    )
+
+    spec = torch.sqrt(spec.pow(2).sum(-1) + 1e-8)
+
+    spec = torch.matmul(mel_basis[fmax_dtype_device], spec)
+    spec = spectral_normalize_torch(spec)
+
+    return spec