o %ݫi@sFdZddlZddlmmZddlmZddZddZdd Z dS) z:Non-negative matrix factorization Authors * Cem Subakan N)spectral_magnitudecCs@t|dddddddf|dddddddf}|S)abReturns the phase of a complex spectrogram. Arguments --------- stft : torch.Tensor A tensor, output from the stft function. Returns ------- phase : torch.Tensor Example ------- >>> BS, nfft, T = 10, 20, 300 >>> X_stft = torch.randn(BS, nfft//2 + 1, T, 2) >>> phase_mix = spectral_phase(X_stft) Nr)torchatan2)stftphaserN/home/ubuntu/.local/lib/python3.10/site-packages/speechbrain/processing/NMF.pyspectral_phase s<r cCs|\}}|jd}|dddd|d}|jd}d}|jdd|}||}tj||gdd} | d} |d} dt| |} | tj| dd|} t dD]!} |t | | |}| t | |}|tj|dd|} qV| |9} t | d d d | f| d | d d f}tj | d|d|dd}tj|dd}t | d d | d f| | d d d f}tj | d|d|dd}tj|dd}||fS) aThis function separates the mixture signals, given NMF template matrices. Arguments --------- Whats : list This list contains the list [W1, W2], where W1 W2 are respectively the NMF template matrices that correspond to source1 and source2. W1, W2 are of size [nfft/2 + 1, K], where nfft is the fft size for STFT, and K is the number of vectors (templates) in W. Xmix : torch.Tensor This is the magnitude spectra for the mixtures. The size is [BS x T x nfft//2 + 1] where, BS = batch size, nfft = fft size, T = number of time steps in the spectra. Returns ------- X1hat : Separated spectrum for source1 Size = [BS x (nfft/2 +1) x T] where, BS = batch size, nfft = fft size, T = number of time steps in the spectra. X2hat : Separated Spectrum for source2 The size definitions are the same as above. Example ------- >>> BS, nfft, T = 4, 20, 400 >>> K1, K2 = 10, 10 >>> W1hat = torch.randn(nfft//2 + 1, K1) >>> W2hat = torch.randn(nfft//2 + 1, K2) >>> Whats = [W1hat, W2hat] >>> Xmix = torch.randn(BS, T, nfft//2 + 1) >>> X1hat, X2hat = NMF_separate_spectra(Whats, Xmix) rrg#B ;dimg?iN) shapepermutereshapesizetsumrcatrandrangematmulsplit unsqueeze)WhatsXmixW1W2 nmixturesnepsgzwKK1hepvnhXhat1Xhat2rrr NMF_separate_spectra$s0!      , , r-c Cstj|||d}t|}t|dd}gg} } d} t|jdD]} || | || || d|| dtjt || dt || dgdd} || | || || d|| dtjt || dt || dgdd}| d dddd } |d dddd }|| }||}d }||| }||| }| || |q | | fS) aThis function reconstructs the separated spectra into waveforms. Arguments --------- X1hat : torch.Tensor The separated spectrum for source 1 of size [BS, nfft/2 + 1, T], where, BS = batch size, nfft = fft size, T = length of the spectra. X2hat : torch.Tensor The separated spectrum for source 2 of size [BS, nfft/2 + 1, T]. The size definitions are the same as Xhat1. X_stft : torch.Tensor This is the magnitude spectra for the mixtures. The size is [BS x nfft//2 + 1 x T x 2] where, BS = batch size, nfft = fft size, T = number of time steps in the spectra. The last dimension is to represent complex numbers. sample_rate : int The sampling rate (in Hz) in which we would like to save the results. win_length : int The length of stft windows (in ms). hop_length : int The length with which we shift the STFT windows (in ms). Returns ------- x1hats : list List of waveforms for source 1. x2hats : list List of waveforms for source 2. Example ------- >>> BS, nfft, T = 10, 512, 16000 >>> sample_rate, win_length, hop_length = 16000, 25, 10 >>> X1hat = torch.randn(BS, nfft//2 + 1, T) >>> X2hat = torch.randn(BS, nfft//2 + 1, T) >>> X_stft = torch.randn(BS, nfft//2 + 1, T, 2) >>> x1hats, x2hats = reconstruct_results(X1hat, X2hat, X_stft, sample_rate, win_length, hop_length) ) sample_rate win_length hop_lengthr )powerg}:rr r r )spfISTFTr rrrrrrcossinrstdappend)X1hatX2hatX_stftr.r/r0r5 phase_mixmag_mixx1hatsx2hatsr!i X1hat_stft X2hat_stftshat1shat2 div_factorx1x2rrr reconstruct_resultsjsP.        rI) __doc__rspeechbrain.processing.features processingfeaturesr4rr r-rIrrrr s  F