core.config

AudioBlock

class fastaudio.core.config.AudioBlock(cache_folder=None, sample_rate=16000, force_mono=True, crop_signal_to=None)

A TransformBlock for audios

from_folder(path, sample_rate=16000, force_mono=True, crop_signal_to=None, **kwargs)

Build a AudioBlock from a path and caches some intermediary results

preprocess_audio_folder

fastaudio.core.config.preprocess_audio_folder(path, folders=None, output_dir=None, sample_rate=16000, force_mono=True, crop_signal_to=None, **kwargs)

Preprocess audio files in path in parallel using n_workers

PreprocessAudio

class fastaudio.core.config.PreprocessAudio(sample_rate=16000, force_mono=True, crop_signal_to=None)

Creates an audio tensor and run the basic preprocessing transforms on it. Used while preprocessing the audios, this is not a Transform.

AudioConfig

class fastaudio.core.config.AudioConfig()

Collection of configurations to build AudioToSpec transforms.

class BasicMFCC(sample_rate=16000, n_mfcc=40, dct_type=2, norm='ortho', log_mels=False, melkwargs=None)

class BasicMelSpectrogram(sample_rate=16000, n_fft=400, win_length=None, hop_length=None, f_min=0.0, f_max=None, pad=0, n_mels=128, window_fn=, power=2.0, normalized=False, wkwargs=None, center=True, pad_mode='reflect', onesided=True, norm=None, mel=True, to_db=True)

class BasicSpectrogram(n_fft=400, win_length=None, hop_length=None, pad=0, window_fn=, power=2.0, normalized=False, wkwargs=None, center=True, pad_mode='reflect', onesided=True, mel=False, to_db=True)

class Voice(sample_rate=16000, n_fft=1024, win_length=None, hop_length=128, f_min=50.0, f_max=8000.0, pad=0, n_mels=128, window_fn=, power=2.0, normalized=False, wkwargs=None, center=True, pad_mode='reflect', onesided=True, norm=None, mel='True', to_db='False')