File size: 5,764 Bytes
43b7e92 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 |
# Copyright 2024 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import numpy as np # noqa: E402
from ....configuration_utils import ConfigMixin, register_to_config
from ....schedulers.scheduling_utils import SchedulerMixin
try:
import librosa # noqa: E402
_librosa_can_be_imported = True
_import_error = ""
except Exception as e:
_librosa_can_be_imported = False
_import_error = (
f"Cannot import librosa because {e}. Make sure to correctly install librosa to be able to install it."
)
from PIL import Image # noqa: E402
class Mel(ConfigMixin, SchedulerMixin):
"""
Parameters:
x_res (`int`):
x resolution of spectrogram (time).
y_res (`int`):
y resolution of spectrogram (frequency bins).
sample_rate (`int`):
Sample rate of audio.
n_fft (`int`):
Number of Fast Fourier Transforms.
hop_length (`int`):
Hop length (a higher number is recommended if `y_res` < 256).
top_db (`int`):
Loudest decibel value.
n_iter (`int`):
Number of iterations for Griffin-Lim Mel inversion.
"""
config_name = "mel_config.json"
@register_to_config
def __init__(
self,
x_res: int = 256,
y_res: int = 256,
sample_rate: int = 22050,
n_fft: int = 2048,
hop_length: int = 512,
top_db: int = 80,
n_iter: int = 32,
):
self.hop_length = hop_length
self.sr = sample_rate
self.n_fft = n_fft
self.top_db = top_db
self.n_iter = n_iter
self.set_resolution(x_res, y_res)
self.audio = None
if not _librosa_can_be_imported:
raise ValueError(_import_error)
def set_resolution(self, x_res: int, y_res: int):
"""Set resolution.
Args:
x_res (`int`):
x resolution of spectrogram (time).
y_res (`int`):
y resolution of spectrogram (frequency bins).
"""
self.x_res = x_res
self.y_res = y_res
self.n_mels = self.y_res
self.slice_size = self.x_res * self.hop_length - 1
def load_audio(self, audio_file: str = None, raw_audio: np.ndarray = None):
"""Load audio.
Args:
audio_file (`str`):
An audio file that must be on disk due to [Librosa](https://librosa.org/) limitation.
raw_audio (`np.ndarray`):
The raw audio file as a NumPy array.
"""
if audio_file is not None:
self.audio, _ = librosa.load(audio_file, mono=True, sr=self.sr)
else:
self.audio = raw_audio
# Pad with silence if necessary.
if len(self.audio) < self.x_res * self.hop_length:
self.audio = np.concatenate([self.audio, np.zeros((self.x_res * self.hop_length - len(self.audio),))])
def get_number_of_slices(self) -> int:
"""Get number of slices in audio.
Returns:
`int`:
Number of spectograms audio can be sliced into.
"""
return len(self.audio) // self.slice_size
def get_audio_slice(self, slice: int = 0) -> np.ndarray:
"""Get slice of audio.
Args:
slice (`int`):
Slice number of audio (out of `get_number_of_slices()`).
Returns:
`np.ndarray`:
The audio slice as a NumPy array.
"""
return self.audio[self.slice_size * slice : self.slice_size * (slice + 1)]
def get_sample_rate(self) -> int:
"""Get sample rate.
Returns:
`int`:
Sample rate of audio.
"""
return self.sr
def audio_slice_to_image(self, slice: int) -> Image.Image:
"""Convert slice of audio to spectrogram.
Args:
slice (`int`):
Slice number of audio to convert (out of `get_number_of_slices()`).
Returns:
`PIL Image`:
A grayscale image of `x_res x y_res`.
"""
S = librosa.feature.melspectrogram(
y=self.get_audio_slice(slice), sr=self.sr, n_fft=self.n_fft, hop_length=self.hop_length, n_mels=self.n_mels
)
log_S = librosa.power_to_db(S, ref=np.max, top_db=self.top_db)
bytedata = (((log_S + self.top_db) * 255 / self.top_db).clip(0, 255) + 0.5).astype(np.uint8)
image = Image.fromarray(bytedata)
return image
def image_to_audio(self, image: Image.Image) -> np.ndarray:
"""Converts spectrogram to audio.
Args:
image (`PIL Image`):
An grayscale image of `x_res x y_res`.
Returns:
audio (`np.ndarray`):
The audio as a NumPy array.
"""
bytedata = np.frombuffer(image.tobytes(), dtype="uint8").reshape((image.height, image.width))
log_S = bytedata.astype("float") * self.top_db / 255 - self.top_db
S = librosa.db_to_power(log_S)
audio = librosa.feature.inverse.mel_to_audio(
S, sr=self.sr, n_fft=self.n_fft, hop_length=self.hop_length, n_iter=self.n_iter
)
return audio
|