Methodforward output = forward(signal, system_ir) signal: (batchsize, length1, dim) system_ir: (length2, dim) output: (batchsize, l
NeuralSeq/modules/parallel_wavegan/models/source.py:216
Methodforward cyc, noise, uv = SourceModuleCycNoise_v1(F0, beta) F0_upsampled (batchsize, length, 1) beta (1) cyc (batchsize, lengt
NeuralSeq/modules/parallel_wavegan/models/source.py:467
Methodforward Sine_source, noise_source = SourceModuleHnNSF(F0_sampled) F0_sampled (batchsize, length, 1) Sine_source (batchsize, length, 1
NeuralSeq/modules/parallel_wavegan/models/source.py:518
Methodforward(self, txt_tokens, word_tokens, ph2word, word_len, mel2word=None, mel2ph=None,
spk_embed=None,
NeuralSeq/modules/syntaspeech/syntaspeech.py:122
Methodforward graph_lst: [list of dgl_graph] ph_encoding: [batch, hid, t_p] ph2word: [list of list[1,2,2,2,3,3,3]] etypes_lst: [lis
NeuralSeq/modules/syntaspeech/syntactic_graph_encoder.py:135
Methodforward :param x: [B, C, T, n_bins] :return: validity: [B, 1], h: List of hiddens
NeuralSeq/modules/syntaspeech/multi_window_disc.py:32
Methodforward Args: x (tensor): input mel, (B, c_in, T, n_bins). x_length (tensor): len of per mel. (B,). Returns:
NeuralSeq/modules/syntaspeech/multi_window_disc.py:55
Methodforward(self, x, x_mask=None, bert_feats=None, ph2word=None, **kwargs)
NeuralSeq/modules/commons/rel_transformer.py:582
Methodforward(self, x, x_mask=None, reverse=False, g=None, **kwargs)
NeuralSeq/modules/commons/normalizing_flow/glow_modules.py:241
Methodforward(self, x, x_mask=None, g=None, reverse=False, return_hiddens=False)
NeuralSeq/modules/commons/normalizing_flow/glow_modules.py:327
Methodforward Args: h_dec: decoder hidden (query), [N, hs_dec] h_enc: encoder memory (key/value), [N, src_max_len, hs_enc]
audio_to_text/captioning/models/decoder.py:124
Methodforward Args: h_dec: decoder hidden (query), [N, hs_dec] h_enc: encoder memory (key/value), [N, src_max_len, hs_enc]
audio_to_text/captioning/models/decoder.py:166