def __init__(self, sample_rate, n_fft, top_db, max_perc):
super().__init__()
self.time_stretch = TimeStretch(hop_length=None, n_freq=n_fft // 2 + 1)
self.stft = Spectrogram(n_fft=n_fft, power=None)
self.com_norm = ComplexNorm(power=2.)
self.mel_specgram = MelSpectrogram(sample_rate,
n_fft=n_fft,
f_max=8000)
self.AtoDB = AmplitudeToDB(top_db=top_db)
self.dist = Uniform(1. - max_perc, 1 + max_perc)
def __init__(self,
df,
sound_dir,
audio_sec=5,
sample_rate=32000,
n_fft=2**11,
top_db=80
):
self.train_df = df
self.sound_dir = sound_dir
self.audio_sec = audio_sec
self.sample_rate = sample_rate
self.mel = MelSpectrogram(sample_rate, n_fft=n_fft)
self.norm_db = AmplitudeToDB(top_db=top_db)
self.time_strech = TimeStretch()
self.target_lenght = sample_rate * audio_sec
def __init__(self, sample_rate, n_fft, top_db, max_perc):
super().__init__()
self.time_stretch = TimeStretch(hop_length=None, n_freq=n_fft//2+1)
self.stft = Spectrogram(n_fft=n_fft, power=None)
self.com_norm = ComplexNorm(power=2.)
self.fm = FrequencyMasking(50)
self.tm = TimeMasking(50)
self.mel_specgram = MelSpectrogram(sample_rate, n_fft=n_fft, f_max=8000)
self.AtoDB= AmplitudeToDB(top_db=top_db)
self.max_perc = max_perc
self.sample_rate = sample_rate
self.resamples = [
Resample(sample_rate, sample_rate*0.6),
Resample(sample_rate, sample_rate*0.7),
Resample(sample_rate, sample_rate*0.8),
Resample(sample_rate, sample_rate*0.9),
Resample(sample_rate, sample_rate*1),
Resample(sample_rate, sample_rate*1.1),
Resample(sample_rate, sample_rate*1.2),
Resample(sample_rate, sample_rate*1.3),
Resample(sample_rate, sample_rate*1.4)
]