Medix-AI/coqui-tts
Medix-AI
/
coqui-tts
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Add pitch_fmin pitch_fmax args to the audio

This commit is contained in:
Eren Gölge 2022-01-25 09:26:23 +00:00
parent 1445a46e9e
commit cd5d1497cf
1 changed files with 13 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
TTS/utils/audio.py
View File

@ -239,6 +239,12 @@ class AudioProcessor(object):
mel_fmax (int, optional): mel_fmax (int, optional):
maximum filter frequency for computing melspectrograms. Defaults to None. maximum filter frequency for computing melspectrograms. Defaults to None.
pitch_fmin (int, optional):
minimum filter frequency for computing pitch. Defaults to None.
pitch_fmax (int, optional):
maximum filter frequency for computing pitch. Defaults to None.
spec_gain (int, optional): spec_gain (int, optional):
gain applied when converting amplitude to DB. Defaults to 20. gain applied when converting amplitude to DB. Defaults to 20.
@ -300,6 +306,8 @@ class AudioProcessor(object):
max_norm=None, max_norm=None,
mel_fmin=None, mel_fmin=None,
mel_fmax=None, mel_fmax=None,
pitch_fmax=None,
pitch_fmin=None,
spec_gain=20, spec_gain=20,
stft_pad_mode="reflect", stft_pad_mode="reflect",
clip_norm=True, clip_norm=True,
@ -333,6 +341,8 @@ class AudioProcessor(object):
self.symmetric_norm = symmetric_norm self.symmetric_norm = symmetric_norm
self.mel_fmin = mel_fmin or 0 self.mel_fmin = mel_fmin or 0
self.mel_fmax = mel_fmax self.mel_fmax = mel_fmax
self.pitch_fmin = pitch_fmin
self.pitch_fmax = pitch_fmax
self.spec_gain = float(spec_gain) self.spec_gain = float(spec_gain)
self.stft_pad_mode = stft_pad_mode self.stft_pad_mode = stft_pad_mode
self.max_norm = 1.0 if max_norm is None else float(max_norm) self.max_norm = 1.0 if max_norm is None else float(max_norm)
@ -726,12 +736,12 @@ class AudioProcessor(object):
>>> WAV_FILE = filename = librosa.util.example_audio_file() >>> WAV_FILE = filename = librosa.util.example_audio_file()
>>> from TTS.config import BaseAudioConfig >>> from TTS.config import BaseAudioConfig
>>> from TTS.utils.audio import AudioProcessor >>> from TTS.utils.audio import AudioProcessor
>>> conf = BaseAudioConfig(mel_fmax=8000) >>> conf = BaseAudioConfig(pitch_fmax=8000)
>>> ap = AudioProcessor(**conf) >>> ap = AudioProcessor(**conf)
>>> wav = ap.load_wav(WAV_FILE, sr=22050)[:5 * 22050] >>> wav = ap.load_wav(WAV_FILE, sr=22050)[:5 * 22050]
>>> pitch = ap.compute_f0(wav) >>> pitch = ap.compute_f0(wav)
""" """
assert self.mel_fmax is not None, " [!] Set `mel_fmax` before caling `compute_f0`." assert self.pitch_fmax is not None, " [!] Set `pitch_fmax` before caling `compute_f0`."
# align F0 length to the spectrogram length # align F0 length to the spectrogram length
if len(x) % self.hop_length == 0: if len(x) % self.hop_length == 0:
x = np.pad(x, (0, self.hop_length // 2), mode="reflect") x = np.pad(x, (0, self.hop_length // 2), mode="reflect")
@ -739,7 +749,7 @@ class AudioProcessor(object):
f0, t = pw.dio( f0, t = pw.dio(
x.astype(np.double), x.astype(np.double),
fs=self.sample_rate, fs=self.sample_rate,
f0_ceil=self.mel_fmax, f0_ceil=self.pitch_fmax,
frame_period=1000 * self.hop_length / self.sample_rate, frame_period=1000 * self.hop_length / self.sample_rate,
) )
f0 = pw.stonemask(x.astype(np.double), f0, t, self.sample_rate) f0 = pw.stonemask(x.astype(np.double), f0, t, self.sample_rate)