File size: 5,602 Bytes
477e9f5
 
08d7176
 
 
 
072ec01
 
08d7176
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6cee2a2
08d7176
6cee2a2
08d7176
 
 
6cee2a2
 
08d7176
 
6cee2a2
08d7176
 
 
 
 
6cee2a2
 
 
 
 
 
 
 
 
 
 
 
 
08d7176
 
 
 
 
 
 
6cee2a2
08d7176
 
 
 
 
 
 
 
 
 
 
 
 
6cee2a2
 
 
08d7176
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
# Adapted from https://github.com/m-bain/whisperX/blob/main/whisperx/audio.py

import os
import subprocess
from functools import lru_cache
from typing import Optional, Union
from scipy.io.wavfile import write
import tempfile

import numpy as np
import torch
import torch.nn.functional as F

def exact_div(x, y):
    assert x % y == 0
    return x // y

# hard-coded audio hyperparameters
SAMPLE_RATE = 16000
N_FFT = 400
HOP_LENGTH = 160
CHUNK_LENGTH = 30
N_SAMPLES = CHUNK_LENGTH * SAMPLE_RATE  # 480000 samples in a 30-second chunk
N_FRAMES = exact_div(N_SAMPLES, HOP_LENGTH)  # 3000 frames in a mel spectrogram input

N_SAMPLES_PER_TOKEN = HOP_LENGTH * 2  # the initial convolutions has stride 2
FRAMES_PER_SECOND = exact_div(SAMPLE_RATE, HOP_LENGTH)  # 10ms per audio frame
TOKENS_PER_SECOND = exact_div(SAMPLE_RATE, N_SAMPLES_PER_TOKEN)  # 20ms per audio token


def load_audio(file: Union[str, np.ndarray], sr: int = SAMPLE_RATE) -> np.ndarray:
    """
    Open an audio file or process a numpy array containing audio data as mono waveform, resampling as necessary.

    Parameters
    ----------
    file: Union[str, np.ndarray]
        The audio file to open or a numpy array containing the audio data.

    sr: int
        The sample rate to resample the audio if necessary.

    Returns
    -------
    A NumPy array containing the audio waveform, in float32 dtype.
    """
    if isinstance(file, np.ndarray):
        if file.dtype != np.float32:
            file = file.astype(np.float32)
        if file.ndim > 1:
            file = np.mean(file, axis=1)

        temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".wav")
        write(temp_file.name, SAMPLE_RATE, (file * 32768).astype(np.int16))
        temp_file_path = temp_file.name
        temp_file.close()
    else:
        temp_file_path = file

    try:
        cmd = [
            "ffmpeg",
            "-nostdin",
            "-threads",
            "0",
            "-i",
            temp_file_path,
            "-f",
            "s16le",
            "-ac",
            "1",
            "-acodec",
            "pcm_s16le",
            "-ar",
            str(sr),
            "-",
        ]
        out = subprocess.run(cmd, capture_output=True, check=True).stdout
    except subprocess.CalledProcessError as e:
        raise RuntimeError(f"Failed to load audio: {e.stderr.decode()}") from e
    finally:
        if isinstance(file, np.ndarray):
            os.remove(temp_file_path)

    return np.frombuffer(out, np.int16).flatten().astype(np.float32) / 32768.0


def pad_or_trim(array, length: int = N_SAMPLES, *, axis: int = -1):
    """
    Pad or trim the audio array to N_SAMPLES, as expected by the encoder.
    """
    if torch.is_tensor(array):
        if array.shape[axis] > length:
            array = array.index_select(
                dim=axis, index=torch.arange(length, device=array.device)
            )

        if array.shape[axis] < length:
            pad_widths = [(0, 0)] * array.ndim
            pad_widths[axis] = (0, length - array.shape[axis])
            array = F.pad(array, [pad for sizes in pad_widths[::-1] for pad in sizes])
    else:
        if array.shape[axis] > length:
            array = array.take(indices=range(length), axis=axis)

        if array.shape[axis] < length:
            pad_widths = [(0, 0)] * array.ndim
            pad_widths[axis] = (0, length - array.shape[axis])
            array = np.pad(array, pad_widths)

    return array


@lru_cache(maxsize=None)
def mel_filters(device, n_mels: int) -> torch.Tensor:
    """
    load the mel filterbank matrix for projecting STFT into a Mel spectrogram.
    Allows decoupling librosa dependency; saved using:

        np.savez_compressed(
            "mel_filters.npz",
            mel_80=librosa.filters.mel(sr=16000, n_fft=400, n_mels=80),
        )
    """
    assert n_mels in [80, 128], f"Unsupported n_mels: {n_mels}"
    with np.load(
        os.path.join(os.path.dirname(__file__), "assets", "mel_filters.npz")
    ) as f:
        return torch.from_numpy(f[f"mel_{n_mels}"]).to(device)


def log_mel_spectrogram(
    audio: Union[str, np.ndarray, torch.Tensor],
    n_mels: int,
    padding: int = 0,
    device: Optional[Union[str, torch.device]] = None,
):
    """
    Compute the log-Mel spectrogram of

    Parameters
    ----------
    audio: Union[str, np.ndarray, torch.Tensor], shape = (*)
        The path to audio or either a NumPy array or Tensor containing the audio waveform in 16 kHz

    n_mels: int
        The number of Mel-frequency filters, only 80 is supported

    padding: int
        Number of zero samples to pad to the right

    device: Optional[Union[str, torch.device]]
        If given, the audio tensor is moved to this device before STFT

    Returns
    -------
    torch.Tensor, shape = (80, n_frames)
        A Tensor that contains the Mel spectrogram
    """
    if not torch.is_tensor(audio):
        if isinstance(audio, str):
            audio = load_audio(audio)
        audio = torch.from_numpy(audio)

    if device is not None:
        audio = audio.to(device)
    if padding > 0:
        audio = F.pad(audio, (0, padding))
    window = torch.hann_window(N_FFT).to(audio.device)
    stft = torch.stft(audio, N_FFT, HOP_LENGTH, window=window, return_complex=True)
    magnitudes = stft[..., :-1].abs() ** 2

    filters = mel_filters(audio.device, n_mels)
    mel_spec = filters @ magnitudes

    log_spec = torch.clamp(mel_spec, min=1e-10).log10()
    log_spec = torch.maximum(log_spec, log_spec.max() - 8.0)
    log_spec = (log_spec + 4.0) / 4.0
    return log_spec