HPSS

class audioflux.HPSS(radix2_exp=12, window_type=WindowType.HAMM, slide_length=1024, h_order=21, p_order=31)

HPSS - Median filtering, NMF algorithm.

Parameters

radix2_exp: int

fft_length=2**radix2_exp

window_type: WindowType

Window type for each frame.

See: type.WindowType

slide_length: int

Window sliding length.

h_order: int

h order

p_order: int

p order

Examples

Get a chord audio file with a metronome

>>> import audioflux as af
>>> audio_arr, sr = af.read(af.utils.sample_path('chord_metronome1'))

Create HPSS object and extrct h/p data

>>> from audioflux.type import WindowType
>>> radix2_exp = 11
>>> slide_length = (1 << radix2_exp) // 4
>>> hpss_obj = af.HPSS(radix2_exp=radix2_exp, window_type=WindowType.HAMM,
>>>                    slide_length=slide_length, h_order=21, p_order=31)
>>> h_arr, p_arr = hpss_obj.hpss(audio_arr)
>>> audio_len = h_arr.shape[-1]

Disable Plot of Linear spectrogram

>>> import numpy as np
>>> from audioflux.type import SpectralFilterBankScaleType, SpectralDataType
>>> bft_obj = af.BFT(num=2049, radix2_exp=12, samplate=sr,
>>>                  scale_type=SpectralFilterBankScaleType.LINEAR,
>>>                  data_type=SpectralDataType.POWER)
>>> audio_arr = audio_arr[..., :audio_len]
>>> origin_spec_arr = bft_obj.bft(audio_arr, result_type=1)
>>> h_spec_arr = bft_obj.bft(h_arr, result_type=1)
>>> p_spec_arr = bft_obj.bft(p_arr, result_type=1)
>>> origin_spec_arr = af.utils.power_to_abs_db(origin_spec_arr)
>>> h_spec_arr = af.utils.power_to_abs_db(h_spec_arr)
>>> p_spec_arr = af.utils.power_to_abs_db(p_spec_arr)

>>> import matplotlib.pyplot as plt
>>> from audioflux.display import fill_spec
>>>
>>> fig, ax = plt.subplots()
>>> img = fill_spec(origin_spec_arr, axes=ax,
>>>           x_coords=bft_obj.x_coords(audio_len),
>>>           y_coords=bft_obj.y_coords(),
>>>           x_axis='time', y_axis='log',
>>>           title='Origin Linear Spectrogram')
>>> fig.colorbar(img, ax=ax)
>>>
>>> fig, ax = plt.subplots()
>>> img = fill_spec(h_spec_arr, axes=ax,
>>>           x_coords=bft_obj.x_coords(audio_len),
>>>           y_coords=bft_obj.y_coords(),
>>>           x_axis='time', y_axis='log',
>>>           title='h_order Linear Spectrogram')
>>> fig.colorbar(img, ax=ax)
>>>
>>> fig, ax = plt.subplots()
>>> img = fill_spec(p_spec_arr, axes=ax,
>>>           x_coords=bft_obj.x_coords(audio_len),
>>>           y_coords=bft_obj.y_coords(),
>>>           x_axis='time', y_axis='log',
>>>           title='p_order Linear Spectrogram')
>>> fig.colorbar(img, ax=ax)

Methods

cal_data_length(data_length)	Calculate the data length.
hpss(data_arr)	Compute the hpss

cal_data_length(data_length)

Calculate the data length.

Parameters

data_length: int

Input array length

Returns

out: int

hpss(data_arr)

Compute the hpss

Parameters

data_arr: np.ndarray [shape=(…, n)]

Audio data array

Returns

h_arr: np.ndarray [shape=(…, n)]

p_arr: np.ndarray [shape=(…, n)]