o i@shddlmZddlmZmZmZmZmZddlZddl m Z ddl m Z ddl mZGddde ZdS) ) OrderedDict)DictListOptionalTupleUnionN) ComplexTensor) AbsSeparator)RNNcseZdZ       ddeded ed ed ed ed edeffdd Z ddeej e fdej de e de eeej e fej effddZeddZZS) DPCLSeparatorblstmtanh( input_dimrnn_typenum_spk nonlinearlayerunitemb_Ddropoutc s~t||_t||||||d|_tj||||_|dvr(t d |tj tj tj d||_||_dS)asDeep Clustering Separator. References: [1] Deep clustering: Discriminative embeddings for segmentation and separation; John R. Hershey. et al., 2016; https://ieeexplore.ieee.org/document/7471631 [2] Manifold-Aware Deep Clustering: Maximizing Angles Between Embedding Vectors Based on Regular Simplex; Tanaka, K. et al., 2021; https://www.isca-speech.org/archive/interspeech_2021/tanaka21_interspeech.html Args: input_dim: input feature dimension rnn_type: string, select from 'blstm', 'lstm' etc. bidirectional: bool, whether the inter-chunk RNN layers are bidirectional. num_spk: number of speakers nonlinear: the nonlinear function for mask estimation, select from 'relu', 'tanh', 'sigmoid' layer: int, number of stacked RNN layers. Default is 3. unit: int, dimension of the hidden state. emb_D: int, dimension of the feature vector for a tf-bin. dropout: float, dropout ratio. Default is 0. )idimelayerscdimhdimrtyp)sigmoidrelurzNot supporting nonlinear={}N)super__init___num_spkr r torchnnLinearlinear ValueErrorformatSigmoidReLUTanhrD) selfrrrrrrrr __class__X/home/ubuntu/.local/lib/python3.10/site-packages/espnet2/enh/separator/dpcl_separator.pyr" s( !  zDPCLSeparator.__init__Ninputilens additionalreturncCst|tr t|}n|}|j\}}}|||\}}} ||}||}||d|j} |j r3d} n| ddd|j ddf } t j ||||j | jd} t j|||| jd} t|j D]"}t j| | dd|ddfdddd| dddd|f<qa| jdd}t ||kdkrn$|}t|D]}t|j D]}| ||||kfjdd| ||f<qqq\||||}g} t|j D] }| |||kqtd | i}| ||fS) aAForward. Args: input (torch.Tensor or ComplexTensor): Encoded feature [B, T, F] ilens (torch.Tensor): input lengths [Batch] additional (Dict or None): other data included in model NOTE: not used in this model Returns: masked (List[Union(torch.Tensor, ComplexTensor)]): [(B, T, N), ...] ilens (torch.Tensor): (B,) others predicted data, e.g. tf_embedding: OrderedDict[ 'tf_embedding': learned embedding of all T-F bins (B, T * F, D), ] N)deviceTr )dimr tf_embedding) isinstancerabsshaper r'rviewr-trainingr#detachr$emptyr8zerosrangesum unsqueezeargminmeanappendr)r.r3r4r5featureBTFx_r;maskedcentersdist last_labelilabelbothersr1r1r2forwardGsF      $  &  zDPCLSeparator.forwardcCs|jSN)r#)r.r1r1r2rszDPCLSeparator.num_spk)r r rr rrrrY)__name__ __module__ __qualname__intstrfloatr"rr$TensorrrrrrrrXpropertyr __classcell__r1r1r/r2r sJ ?  Ar ) collectionsrtypingrrrrrr$torch_complex.tensorr#espnet2.enh.separator.abs_separatorr (espnet.nets.pytorch_backend.rnn.encodersr r r1r1r1r2s