MCPcopy Create free account

hub / github.com/AIFSH/ComfyUI-GPT_SoVITS / functions

Functions666 in github.com/AIFSH/ComfyUI-GPT_SoVITS

⨍ Functions 666◇ Types & classes 132↳ Endpoints 1
↓ 20 callersMethodkeys
(self)
GPT_SoVITS/utils.py:341
↓ 18 callersMethod__init__
( self, hidden_channels, kernel_size, dilation_rate, n_layers,
GPT_SoVITS/module/modules.py:136
↓ 18 callersFunctionget_padding
(kernel_size, dilation=1)
GPT_SoVITS/module/commons.py:12
↓ 16 callersMethoditems
(self)
GPT_SoVITS/utils.py:344
↓ 14 callersMethod__init__
( self, in_channels, out_channels, hidden_channels, kernel_size,
GPT_SoVITS/module/models.py:359
↓ 14 callersMethod__init__
( self, in_channels, out_channels, hidden_channels, kernel_size,
GPT_SoVITS/module/models_onnx.py:353
↓ 11 callersFunctionnum2str
(value_string: str)
GPT_SoVITS/text/zh_normalization/num.py:233
↓ 10 callersMethodstep
Performs a single optimization step. Arguments: closure (callable, optional): A closure that reevaluates the model
GPT_SoVITS/AR/modules/optim.py:207
↓ 7 callersMethod__init__
( self, in_channels, out_channels, filter_channels, kernel_size,
GPT_SoVITS/module/attentions.py:378
↓ 7 callersMethod_all_tone_three
(self, finals: List[str])
GPT_SoVITS/text/tone_sandhi.py:640
↓ 7 callersFunctioncleaned_text_to_sequence
Converts a string of text to a sequence of IDs corresponding to the symbols in the text. Args: text: string to convert to a sequence Ret
GPT_SoVITS/text/__init__.py:6
↓ 6 callersFunction_numeric_feature_by_regex
(regex, s)
GPT_SoVITS/text/japanese.py:176
↓ 6 callersMethodexport
(self, ref_seq, text_seq, ref_bert, text_bert, ssl_content, project_name, dynamo=False)
GPT_SoVITS/onnx_export.py:132
↓ 6 callersMethodupdate_input_paths
(self, input_box, file_path)
GPT_SoVITS/inference_gui.py:293
↓ 6 callersFunctionverbalize_digit
(value_string: str, alt_one=False)
GPT_SoVITS/text/zh_normalization/num.py:225
↓ 5 callersMethod__init__
(self, t2s_path, vits_model)
GPT_SoVITS/onnx_export.py:87
↓ 5 callersFunctionclean_text
(text, language)
GPT_SoVITS/text/cleaner.py:12
↓ 5 callersMethoddecode
(self, codes, text, refer, noise_scale=0.5)
GPT_SoVITS/module/models.py:962
↓ 5 callersMethodembedding
(self, index: int)
GPT_SoVITS/AR/modules/embedding.py:27
↓ 5 callersFunctionverbalize_cardinal
(value_string: str)
GPT_SoVITS/text/zh_normalization/num.py:208
↓ 4 callersMethod__init__
( self, normalized_shape: _shape_t, eps: float = 1e-5, elementwise_affine: boo
GPT_SoVITS/AR/modules/transformer_onnx.py:28
↓ 4 callersMethod__init__
( self, normalized_shape: _shape_t, eps: float = 1e-5, elementwise_affine: boo
GPT_SoVITS/AR/modules/transformer.py:28
↓ 4 callersFunction_time_num2str
A special case for verbalizing number in time.
GPT_SoVITS/text/zh_normalization/chronology.py:22
↓ 4 callersMethodextract_latent
(self, x)
GPT_SoVITS/module/models.py:249
↓ 4 callersFunctionmake_pad_mask
Args: lengths: A 1-D tensor containing sentence lengths. max_len: The length of masks. Returns: Return a 2-
GPT_SoVITS/AR/models/utils.py:14
↓ 4 callersMethodupdate_ref_audio
(self, file_path)
GPT_SoVITS/inference_gui.py:290
↓ 4 callersMethodweight_norm
(self)
GPT_SoVITS/module/attentions.py:479
↓ 3 callersMethod__init__
(self, ar_text_embedding, bert_proj, ar_text_position)
GPT_SoVITS/AR/models/t2s_model_onnx.py:89
↓ 3 callersMethod__init__
(self, hidden_channels, kernel_size, dilation_rate, n_layers)
GPT_SoVITS/module/mrte_model.py:117
↓ 3 callersMethod__init__
( self, in_channels, out_channels, filter_channels, kernel_size,
GPT_SoVITS/module/attentions_onnx.py:300
↓ 3 callersMethod__setitem__
(self, key, value)
GPT_SoVITS/utils.py:356
↓ 3 callersMethoddecode
(self, embed_ind)
GPT_SoVITS/module/core_vq.py:202
↓ 3 callersFunctiong2p
(text)
GPT_SoVITS/text/english.py:351
↓ 3 callersFunctionget_files
(end_with="pth",model_type="D")
nodes.py:264
↓ 3 callersFunctionkill_process
(pid)
finetune.py:182
↓ 3 callersFunctionload_audio
(file, sr)
GPT_SoVITS/my_utils.py:5
↓ 3 callersFunctionpreviewAudio
(node,file)
web/js/uploadAudio.js:10
↓ 3 callersFunctionsample
( logits, previous_tokens, **sampling_kwargs, )
GPT_SoVITS/AR/models/t2s_model_onnx.py:76
↓ 3 callersFunctionspectrogram_torch
(y, n_fft, sampling_rate, hop_size, win_size, center=False)
GPT_SoVITS/module/mel_processing.py:51
↓ 2 callersFunctionBalancedDoubleSwish
ActivationBalancer -> DoubleSwish
GPT_SoVITS/AR/modules/scaling.py:323
↓ 2 callersMethod__init__
( self, dim: int, codebook_size: int, kmeans_init: int = False, kmeans
GPT_SoVITS/module/core_vq.py:112
↓ 2 callersMethod_ff_block
(self, x: Tensor)
GPT_SoVITS/AR/modules/transformer.py:344
↓ 2 callersFunction_g2p
(segments)
GPT_SoVITS/text/chinese.py:77
↓ 2 callersMethod_get_relative_embeddings
(self, relative_embeddings, length)
GPT_SoVITS/module/attentions.py:310
↓ 2 callersMethod_get_relative_embeddings
(self, relative_embeddings, length)
GPT_SoVITS/module/attentions_onnx.py:232
↓ 2 callersMethod_is_reduplication
(self, word: str)
GPT_SoVITS/text/tone_sandhi.py:730
↓ 2 callersMethod_sa_block
( self, x: Tensor, attn_mask: Optional[Tensor], key_padding_mask: Optional[Ten
GPT_SoVITS/AR/modules/transformer.py:321
↓ 2 callersMethod_split_word
(self, word: str)
GPT_SoVITS/text/tone_sandhi.py:587
↓ 2 callersMethodattention
(self, query, key, value, mask=None)
GPT_SoVITS/module/attentions.py:241
↓ 2 callersMethodbackward
(ctx, y_grad: Tensor)
GPT_SoVITS/AR/modules/scaling.py:78
↓ 2 callersFunctionbatch_sequences
(sequences: List[np.array], axis: int = 0, pad_value: int = 0)
GPT_SoVITS/AR/data/dataset.py:23
↓ 2 callersFunctionchange_gpt_weights
(gpt_path)
GPT_SoVITS/inference_webui.py:165
↓ 2 callersFunctionchange_sovits_weights
(sovits_path)
GPT_SoVITS/inference_webui.py:138
↓ 2 callersFunctionclean_text_inf
(text, language)
inference.py:278
↓ 2 callersFunctionclean_text_inf
(text, language)
GPT_SoVITS/inference_webui.py:211
↓ 2 callersFunctionconvert_pad_shape
(pad_shape)
GPT_SoVITS/module/commons.py:16
↓ 2 callersMethoddequantize
(self, embed_ind)
GPT_SoVITS/module/core_vq.py:188
↓ 2 callersFunctionema_inplace
(moving_avg, new, decay: float)
GPT_SoVITS/module/core_vq.py:46
↓ 2 callersMethodencode
(self, x)
GPT_SoVITS/module/core_vq.py:192
↓ 2 callersMethodextend_pe
Reset the positional encodings.
GPT_SoVITS/AR/modules/embedding.py:54
↓ 2 callersFunctionfitHeight
(node)
web/js/previewAudio.js:4
↓ 2 callersFunctionfitHeight
(node)
web/js/uploadAudio.js:5
↓ 2 callersFunctionget_bert_feature
(text, word2ph)
inference.py:283
↓ 2 callersFunctionget_bert_feature
(text, word2ph)
GPT_SoVITS/inference_webui.py:87
↓ 2 callersFunctionget_phones_and_bert
(text,language)
inference.py:319
↓ 2 callersFunctionget_phones_and_bert
(text,language)
GPT_SoVITS/inference_webui.py:239
↓ 2 callersFunctionget_timing_signal_1d
(length, channels, min_timescale=1.0, max_timescale=1.0e4)
GPT_SoVITS/module/commons.py:67
↓ 2 callersFunctionget_tts_wav
(ref_wav_path, prompt_text, prompt_language, text, text_language, how_to_cut=i18n("不切"), top_
inference.py:125
↓ 2 callersFunctionget_weights_names
()
GPT_SoVITS/inference_webui.py:539
↓ 2 callersMethodinfer
(self, ssl, y, y_lengths, text, text_lengths, test=None, noise_scale=0.5)
GPT_SoVITS/module/models.py:938
↓ 2 callersMethodinfer_panel
( self, x, #####全部文本token x_lens, prompts, ####参考音频token bert_featur
GPT_SoVITS/AR/models/t2s_model.py:324
↓ 2 callersMethodmake_input_data
(self, x, x_lens, y, y_lens, bert_feature)
GPT_SoVITS/AR/models/t2s_model.py:92
↓ 2 callersFunctionmel_spectrogram_torch
( y, n_fft, num_mels, sampling_rate, hop_size, win_size, fmin, fmax, center=False )
GPT_SoVITS/module/mel_processing.py:104
↓ 2 callersFunctionname2go
(wav_name,wav_path)
GPT_SoVITS/prepare_datasets/2-get-hubert-wav32k.py:65
↓ 2 callersMethodnormalize
(self, text: str)
GPT_SoVITS/text/zh_normalization/text_normlization.py:155
↓ 2 callersMethodpad_y_eos
(self, y, y_mask_int, eos_id)
GPT_SoVITS/AR/models/t2s_model.py:317
↓ 2 callersFunctionphone2str
(phone_string: str, mobile=True)
GPT_SoVITS/text/zh_normalization/phonecode.py:33
↓ 2 callersMethodpostprocess_emb
(self, embed_ind, shape)
GPT_SoVITS/module/core_vq.py:185
↓ 2 callersMethodpreprocess
(self, x)
GPT_SoVITS/module/core_vq.py:171
↓ 2 callersMethodquantize
(self, x)
GPT_SoVITS/module/core_vq.py:175
↓ 2 callersFunctionsample_vectors
(samples, num: int)
GPT_SoVITS/module/core_vq.py:60
↓ 2 callersFunctionsearchsorted
(bin_locations, inputs, eps=1e-6)
GPT_SoVITS/module/transforms.py:45
↓ 2 callersFunctionspec_to_mel_torch
(spec, n_fft, num_mels, sampling_rate, fmin, fmax)
GPT_SoVITS/module/mel_processing.py:88
↓ 2 callersFunctionspectral_normalize_torch
(magnitudes)
GPT_SoVITS/module/mel_processing.py:37
↓ 2 callersFunctionsplit
(todo_text)
inference.py:393
↓ 2 callersFunctionsplit
(todo_text)
GPT_SoVITS/inference_webui.py:436
↓ 2 callersFunctiontext_normalize
(text)
GPT_SoVITS/text/english.py:214
↓ 2 callersFunctiontrain_and_evaluate
( rank, epoch, hps, nets, optims, schedulers, scaler, loaders, logger, writers )
GPT_SoVITS/s2_train.py:285
↓ 2 callersFunctiontranditional_to_simplified
(text: str)
GPT_SoVITS/text/zh_normalization/char_convert.py:30
↓ 2 callersMethodtransform
(self, phonemes)
GPT_SoVITS/AR/text_processing/phonemizer.py:65
↓ 2 callersFunctionweight_norm_modules
(module, name="weight", dim=0)
GPT_SoVITS/module/attentions.py:539
↓ 1 callersMethod__init__
( self, params, lr=3e-02, clipping_scale=None, bet
GPT_SoVITS/AR/modules/optim.py:166
↓ 1 callersMethod__init__
( self, embedding_dim: int, vocab_size: int, dropout: float = 0.0, )
GPT_SoVITS/AR/modules/embedding.py:9
↓ 1 callersMethod__init__
( self, embedding_dim: int, vocab_size: int, dropout: float = 0.0, )
GPT_SoVITS/AR/modules/embedding_onnx.py:9
↓ 1 callersMethod__init__
(self, hparams, val=False)
GPT_SoVITS/module/data_utils.py:30
↓ 1 callersMethod__len__
(self)
GPT_SoVITS/AR/data/dataset.py:215
↓ 1 callersMethod_absolute_position_to_relative_position
x: [b, h, l, l] ret: [b, h, l, 2*l-1]
GPT_SoVITS/module/attentions.py:349
↓ 1 callersMethod_absolute_position_to_relative_position
x: [b, h, l, l] ret: [b, h, l, 2*l-1]
GPT_SoVITS/module/attentions_onnx.py:271
↓ 1 callersMethod_attention_bias_proximal
Bias for self-attention to encourage attention to close positions. Args: length: an integer scalar. Returns: a Ten
GPT_SoVITS/module/attentions.py:365
↓ 1 callersMethod_bisect
(self, x, lo=0, hi=None)
GPT_SoVITS/module/data_utils.py:316
next →1–100 of 666, ranked by callers