MCPcopy Create free account

hub / github.com/MoonInTheRiver/DiffSinger / functions

Functions694 in github.com/MoonInTheRiver/DiffSinger

⨍ Functions 694◇ Types & classes 154
↓ 1 callersMethodconfigure_ddp
(self, model, device_ids)
tasks/base_task.py:260
↓ 1 callersFunctionconvert_continuos_f0
CONVERT F0 TO CONTINUOUS F0 Args: f0 (ndarray): original f0 sequence with the shape (T) Return: (ndarray): continuous f0 with
utils/cwt.py:12
↓ 1 callersFunctioncorrect_symbols
一百八 to 一百八十 一亿一千三百万 to 一亿 一千万 三百万
utils/text_norm.py:254
↓ 1 callersFunctioncosine_beta_schedule
cosine schedule as proposed in https://openreview.net/forum?id=-NEXDKk8gZ
usr/diff/diffusion.py:167
↓ 1 callersFunctioncosine_beta_schedule
cosine schedule as proposed in https://openreview.net/forum?id=-NEXDKk8gZ
usr/diff/shallow_diffusion_tts.py:175
↓ 1 callersMethodcwt_loss
(self, cwt_p, cwt_g)
tasks/tts/fs2.py:271
↓ 1 callersMethoddate2chntext
(self)
utils/text_norm.py:536
↓ 1 callersMethoddecode_list
Transform a sequence of int ids into a their string versions. This method supports transforming individual input/output ids to their
utils/text_encoder.py:79
↓ 1 callersMethoddecode_list
(self, ids)
utils/text_encoder.py:216
↓ 1 callersFunctiondenoise
(wav, v=0.1)
vocoders/vocoder_utils.py:7
↓ 1 callersMethoddenorm_spec
(self, x)
usr/diff/diffusion.py:323
↓ 1 callersFunctiondesign_prototype_filter
Design prototype filter for PQMF. This method is based on `A Kaiser window approach for the design of prototype filters of cosine modulated f
modules/parallel_wavegan/layers/pqmf.py:15
↓ 1 callersMethoddump_checkpoint
(self)
utils/pl_utils.py:743
↓ 1 callersFunctiondur_to_figure
(dur_gt, dur_pred, txt)
utils/plot.py:30
↓ 1 callersFunctiondynamic_range_compression_torch
(x, C=1, clip_val=1e-5)
modules/hifigan/mel_utils.py:23
↓ 1 callersFunctiondynamic_range_decompression_torch
(x, C=1)
modules/hifigan/mel_utils.py:27
↓ 1 callersMethodevaluation_forward
(self, model, batch, batch_idx, dataloader_idx, test=False)
utils/pl_utils.py:1281
↓ 1 callersFunctionf0_to_figure
(f0_gt, mel, i, f0_cwt=None, f0_pred=None)
usr/singingvocoder/highgan.py:72
↓ 1 callersMethodforward
Apply layer normalization. :param torch.Tensor x: input tensor :return: layer normalized tensor :rtype torch.Tensor
modules/fastspeech/tts_modules.py:48
↓ 1 callersMethodforward_embedding
(self, txt_tokens)
modules/fastspeech/tts_modules.py:337
↓ 1 callersMethodfraction2chntext
(self)
utils/text_norm.py:498
↓ 1 callersFunctiongaussian
(window_size, sigma)
modules/commons/ssim.py:319
↓ 1 callersFunctionget_a_var
(obj)
utils/pl_utils.py:32
↓ 1 callersMethodget_align
(tg_fn, ph, mel, phone_encoded, res)
data_gen/tts/base_binarizer.py:188
↓ 1 callersMethodget_all_ckpts
(self)
utils/pl_utils.py:323
↓ 1 callersMethodget_dataloaders
(self, model)
utils/pl_utils.py:1096
↓ 1 callersMethodget_f0cwt
(f0, res)
data_gen/tts/base_binarizer.py:208
↓ 1 callersMethodget_pitch
(wav, mel, res)
data_gen/tts/base_binarizer.py:200
↓ 1 callersFunctionget_value
(value_string, use_zeros=True)
utils/text_norm.py:322
↓ 1 callersMethodgrad_norm
(self, norm_type)
tasks/base_task.py:340
↓ 1 callersMethodin_proj_qkv
(self, query)
modules/commons/common_layers.py:422
↓ 1 callersMethodinference
Inference duration. Args: xs (Tensor): Batch of input sequences (B, Tmax, idim). x_masks (ByteTensor, optional): Batch
modules/fastspeech/tts_modules.py:143
↓ 1 callersMethodinit_ddp_connection
(self, proc_rank, world_size)
tasks/base_task.py:276
↓ 1 callersMethodinit_test_dataloader
(self, model)
utils/pl_utils.py:1137
↓ 1 callersMethodinit_train_dataloader
(self, model)
utils/pl_utils.py:1111
↓ 1 callersMethodinit_val_dataloader
(self, model)
utils/pl_utils.py:1127
↓ 1 callersFunctioninverse_cwt
(Wavelet_lf0, scales)
utils/cwt.py:127
↓ 1 callersFunctioninverse_cwt_torch
(Wavelet_lf0, scales)
utils/cwt.py:118
↓ 1 callersMethoditem_name2spk_id
(self, item_name)
data_gen/tts/base_binarizer.py:80
↓ 1 callersFunctionload_config
(config_fn)
utils/hparams.py:47
↓ 1 callersMethodload_meta_data
(self)
data_gen/singing/binarize.py:47
↓ 1 callersFunctionload_model_config
(checkpoint_path, config_path)
usr/singingvocoder/highgan.py:22
↓ 1 callersMethodload_test_inputs
(self, test_input_dir, spk_id=0)
tasks/tts/fs2_utils.py:154
↓ 1 callersMethodmeta_data
(self, prefix)
data_gen/tts/base_binarizer.py:96
↓ 1 callersMethodmetrics_to_scalars
(self, metrics)
utils/pl_utils.py:945
↓ 1 callersMethodmoney2chntext
(self)
utils/text_norm.py:573
↓ 1 callersFunctionmoving_average
(array, width)
data_gen/tts/data_gen_utils.py:75
↓ 1 callersMethodnoise_decay
decayed_noise = noise_decay(beta, f0mean) decayed_noise = n[t]exp(-t * f_mean / beta / samp_rate) beta: (dim=1) or (batchsize=1, 1,
modules/parallel_wavegan/models/source.py:264
↓ 1 callersFunctionnoise_like
(shape, device, repeat=False)
usr/diff/diffusion.py:161
↓ 1 callersFunctionnoise_like
(shape, device, repeat=False)
usr/diff/shallow_diffusion_tts.py:161
↓ 1 callersMethodnorm_spec
(self, x)
usr/diff/diffusion.py:320
↓ 1 callersFunctionnormalize
(S)
usr/singingvocoder/highgan.py:18
↓ 1 callersFunctionnum_params
(model, print_out=True, model_name="model")
utils/__init__.py:245
↓ 1 callersMethodon_after_backward
(self)
tasks/base_task.py:334
↓ 1 callersMethodon_batch_end
(self)
tasks/base_task.py:322
↓ 1 callersMethodon_batch_start
(self, batch)
tasks/base_task.py:319
↓ 1 callersMethodon_epoch_begin
(self, epoch, trainer)
utils/pl_utils.py:276
↓ 1 callersMethodon_epoch_start
(self)
tasks/base_task.py:114
↓ 1 callersMethodon_post_performance_check
(self)
tasks/base_task.py:328
↓ 1 callersMethodon_pre_performance_check
(self)
tasks/base_task.py:325
↓ 1 callersMethodon_sanity_check_start
(self)
tasks/base_task.py:310
↓ 1 callersMethodon_save_checkpoint
(self, checkpoint)
tasks/base_task.py:307
↓ 1 callersMethodon_train_end
(self)
tasks/base_task.py:316
↓ 1 callersMethodon_train_start
(self)
tasks/base_task.py:313
↓ 1 callersMethodoptimizer_step
(self, epoch, batch_idx, optimizer, optimizer_idx)
usr/task.py:78
↓ 1 callersMethodordered_indices
Return an ordered list of indices. Batches will be constructed based on this order.
tasks/base_task.py:60
↓ 1 callersMethodout2mel
(self, x)
usr/diff/diffusion.py:329
↓ 1 callersMethodp_losses
(self, x_start, t, cond, noise=None, nonpadding=None)
usr/diff/diffusion.py:276
↓ 1 callersMethodp_mean_variance
(self, x, t, cond, clip_denoised: bool)
usr/diff/diffusion.py:250
↓ 1 callersMethodp_mean_variance
(self, x, t, cond, clip_denoised: bool)
usr/diff/shallow_diffusion_tts.py:267
↓ 1 callersMethodp_sample
(self, x, t, cond, clip_denoised=True, repeat_noise=False)
usr/diff/diffusion.py:261
↓ 1 callersMethodparallel_apply
(self, replicas, inputs, kwargs)
utils/pl_utils.py:184
↓ 1 callersMethodparallel_apply
(self, replicas, inputs, kwargs)
utils/pl_utils.py:253
↓ 1 callersMethodpercentage2chntext
(self)
utils/text_norm.py:596
↓ 1 callersMethodplot_dur
(self, batch_idx, sample, model_out)
tasks/tts/fs2.py:296
↓ 1 callersMethodplot_pitch
(self, batch_idx, sample, model_out)
tasks/tts/fs2.py:305
↓ 1 callersMethodplot_pitch
(self, batch_idx, model_out, sample)
tasks/tts/pe.py:139
↓ 1 callersMethodpredict_start_from_noise
(self, x_t, t, noise)
usr/diff/diffusion.py:235
↓ 1 callersMethodpredict_start_from_noise
(self, x_t, t, noise)
usr/diff/shallow_diffusion_tts.py:252
↓ 1 callersMethodpreprocess_text
(text)
data_gen/tts/txt_processors/zh.py:14
↓ 1 callersMethodprint_nan_gradients
(self)
utils/pl_utils.py:1079
↓ 1 callersMethodprocess
(self)
data_gen/singing/binarize.py:86
↓ 1 callersMethodprocess
(cls, txt, pre_align_args)
data_gen/tts/txt_processors/en.py:65
↓ 1 callersMethodprocess
(cls, txt, pre_align_args)
data_gen/tts/txt_processors/zh_g2pM.py:20
↓ 1 callersMethodprocess_item
(cls, item_name, ph, txt, tg_fn, wav_fn, spk_id, encoder, binarization_args)
data_gen/tts/base_binarizer.py:160
↓ 1 callersFunctionprocess_utterance
(wav_path, fft_size=1024, hop_size=256, win_
data_gen/tts/data_gen_utils.py:93
↓ 1 callersMethodq_posterior
(self, x_start, x_t, t)
usr/diff/diffusion.py:241
↓ 1 callersMethodq_posterior
(self, x_start, x_t, t)
usr/diff/shallow_diffusion_tts.py:258
↓ 1 callersMethodq_sample
(self, x_start, t, noise=None)
usr/diff/diffusion.py:269
↓ 1 callersMethodremove_weight_norm
(self)
modules/hifigan/hifigan.py:171
↓ 1 callersMethodreset
(self)
utils/__init__.py:33
↓ 1 callersMethodreset_parameters
Reset parameters. This initialization follows official implementation manner. https://github.com/descriptinc/melgan-neurips/blob/mast
modules/parallel_wavegan/models/melgan.py:179
↓ 1 callersMethodreset_parameters
Reset parameters. This initialization follows official implementation manner. https://github.com/descriptinc/melgan-neurips/blob/mast
modules/parallel_wavegan/models/melgan.py:415
↓ 1 callersMethodreset_parameters
(self)
modules/commons/common_layers.py:219
↓ 1 callersMethodresolve_root_node_address
(self, root_node)
utils/pl_utils.py:905
↓ 1 callersMethodrestore_state_if_checkpoint_exists
(self, model)
utils/pl_utils.py:628
↓ 1 callersMethodrestore_training_state
Restore trainer state. Model will get its change to update :param checkpoint: :return:
utils/pl_utils.py:683
↓ 1 callersMethodrestore_weights
To restore weights we have two cases. First, attempt to restore hpc weights. If successful, don't restore other weights.
utils/pl_utils.py:599
↓ 1 callersMethodrun_decoder
(self, decoder_inp, tgt_nonpadding, ret, infer, **kwargs)
modules/fastspeech/fs2.py:233
↓ 1 callersMethodrun_model
(self, model, sample, return_output=False, infer=False)
usr/diffspeech_task.py:48
← previousnext →201–300 of 694, ranked by callers