Functions — github.com/MeiGen-AI/MultiTalk

Methodadd_noise

( self, original_samples: torch.Tensor, noise: torch.Tensor, timesteps: torch.

wan/utils/fm_solvers.py:817

Methodadd_noise

( self, original_samples: torch.Tensor, noise: torch.Tensor, timesteps: torch.

wan/utils/fm_solvers_unipc.py:760

Functionapply_lora

(model, device_to, transformer_load_device, params_to_keep=None, dtype=None, base_dtype=None, state_dict=None,

wan/utils/utils.py:361

Functionattention

( q, k, v, q_lens=None, k_lens=None, dropout_p=0., softmax_scale=None, q_scale

wan/modules/attention.py:142

Methodaudio

(self)

kokoro/pipeline.py:344

Methodbegin_index

The index for the first timestep. It should be set from pipeline with `set_begin_index` method.

wan/utils/fm_solvers.py:211

Methodbegin_index

The index for the first timestep. It should be set from pipeline with `set_begin_index` method.

wan/utils/fm_solvers_unipc.py:144

Functioncache_image

(tensor, save_file, nrow=8, normalize=True, va

wan/utils/utils.py:64

Functioncache_video

(tensor, save_file=None, fps=30, suffix='.mp4',

wan/utils/utils.py:23

Methodcross_attn_ffn

(x, context, context_lens, e)

wan/modules/model.py:309

Methodencode

( self, extract_features, attention_mask=None, mask_time_indices=None,

src/audio_analysis/wav2vec2.py:78

Methodensure_image

(_image)

wan/utils/prompt_extend.py:297

Methodextend

(self, prompt, system_prompt, seed=-1, *args, **kwargs)

wan/utils/prompt_extend.py:252

Methodextend

(self, prompt, system_prompt, seed=-1, *args, **kwargs)

wan/utils/prompt_extend.py:433

Methodextend_with_img

(self, prompt, system_prompt, image: U

wan/utils/prompt_extend.py:288

Methodextend_with_img

(self, prompt, system_prompt, image: U

wan/utils/prompt_extend.py:464

Methodfeature_extract

( self, input_values, seq_len, )

src/audio_analysis/wav2vec2.py:67

Methodforward

(self, x)

wan/modules/clip.py:43

Methodforward

(self, x)

wan/modules/clip.py:49

Methodforward

x: [B, L, C].

wan/modules/clip.py:74

Methodforward

(self, x)

wan/modules/clip.py:146

Methodforward

x: [B, L, C].

wan/modules/clip.py:186

Methodforward

(self, x, interpolation=False, use_31_block=False)

wan/modules/clip.py:279

Methodforward

(self, ids)

wan/modules/clip.py:315

Methodforward

imgs: [B, 3, H, W] of torch.float32. - mean: [0.48145466, 0.4578275, 0.40821073] - std: [0.26862954, 0.2613025

wan/modules/clip.py:406

Methodforward

(self, x)

wan/modules/t5.py:66

Methodforward

x: [B, L1, C]. context: [B, L2, C] or None. mask: [B, L2] or [B, L1, L2] or None.

wan/modules/t5.py:91

Methodforward

(self, x)

wan/modules/t5.py:141

Methodforward

(self, x, mask=None, pos_bias=None)

wan/modules/t5.py:175

Methodforward

(self, x, mask=None, encoder_states=None, enco

wan/modules/t5.py:211

Methodforward

(self, lq, lk)

wan/modules/t5.py:238

Methodforward

(self, ids, mask=None)

wan/modules/t5.py:308

Methodforward

(self, ids, mask=None, encoder_states=None, encoder_mask=None)

wan/modules/t5.py:356

Methodforward

(self, encoder_ids, encoder_mask, decoder_ids, decoder_mask)

wan/modules/t5.py:413

Methodforward

(self, x, cache_x=None)

wan/modules/vae.py:28

Methodforward

(self, x)

wan/modules/vae.py:51

Methodforward

Fix bfloat16 support for nearest neighbor interpolation.

wan/modules/vae.py:59

Methodforward

(self, x, feat_cache=None, feat_idx=[0])

wan/modules/vae.py:101

Methodforward

(self, x, feat_cache=None, feat_idx=[0])

wan/modules/vae.py:202

Methodforward

(self, x)

wan/modules/vae.py:240

Methodforward

(self, x, feat_cache=None, feat_idx=[0])

wan/modules/vae.py:318

Methodforward

(self, x, feat_cache=None, feat_idx=[0])

wan/modules/vae.py:423

Methodforward

r""" Args: x(Tensor): Shape [B, L, C]

wan/modules/multitalk_model.py:87

Methodforward

(self, inputs: torch.Tensor)

wan/modules/multitalk_model.py:103

Methodforward

(self, x, seq_lens, grid_sizes, freqs, ref_target_masks=None)

wan/modules/multitalk_model.py:140

Methodforward

(self, x, context, context_lens)

wan/modules/multitalk_model.py:189

Methodforward

( self, x, e, seq_lens, grid_sizes, freqs, context,

wan/modules/multitalk_model.py:274

Methodforward

r""" Args: x(Tensor): Shape [B, L1, C] e(Tensor): Shape [B, C]

wan/modules/multitalk_model.py:338

Methodforward

(self, image_embeds)

wan/modules/multitalk_model.py:361

Methodforward

(self, audio_embeds, audio_embeds_vf)

wan/modules/multitalk_model.py:396

Methodforward

( self, x, t, context, seq_len, clip_f

wan/modules/multitalk_model.py:597

Methodforward

(self, c, x, **kwargs)

wan/modules/vace_model.py:33

Methodforward

(self, x, hints, context_scale=1.0, **kwargs)

wan/modules/vace_model.py:58

Methodforward

(self, x: torch.Tensor, encoder_hidden_states: torch.Tensor,

wan/modules/attention.py:316

Methodforward

x: [B, L, C].

wan/modules/xlm_roberta.py:27

Methodforward

(self, x, mask)

wan/modules/xlm_roberta.py:66

Methodforward

ids: [B, L] of torch.LongTensor.

wan/modules/xlm_roberta.py:118

Methodforward

r""" Args: x(Tensor): Shape [B, L, C]

wan/modules/model.py:81

Methodforward

r""" Args: x(Tensor): Shape [B, L, C]

wan/modules/model.py:97

Methodforward

r""" Args: x(Tensor): Shape [B, L, num_heads, C / num_heads] seq_lens(Tensor): Shape [B] grid_sizes(Tensor

wan/modules/model.py:130

Methodforward

r""" Args: x(Tensor): Shape [B, L1, C] context(Tensor): Shape [B, L2, C] context_lens(Tensor): Shape [B]

wan/modules/model.py:164

Methodforward

r""" Args: x(Tensor): Shape [B, L1, C] context(Tensor): Shape [B, L2, C] context_lens(Tensor): Shape [B]

wan/modules/model.py:202

Methodforward

r""" Args: x(Tensor): Shape [B, L, C] e(Tensor): Shape [B, 6, C] seq_lens(Tensor): Shape [B], length of ea

wan/modules/model.py:278

Methodforward

(self, image_embeds)

wan/modules/model.py:363

Methodforward

r""" Forward pass through the diffusion model Args: x (List[Tensor]): List of input video tensors, each w

wan/modules/model.py:493

Methodforward

1D RoPE. Args: query (torch.tensor): [B, head, seq, head_dim] pos_indices (torch.tensor): [seq,] Returns:

wan/utils/multitalk_utils.py:173

Methodforward

(self, x, s)

kokoro/istftnet.py:27

Methodforward

(self, x, s)

kokoro/istftnet.py:68

Methodforward

(self, input_data)

kokoro/istftnet.py:102

Methodforward

sine_tensor, uv = forward(f0) input F0: tensor(batchsize=1, length, dim=1) f0 for unvoiced steps should be 0 output

kokoro/istftnet.py:185

Methodforward

Sine_source, noise_source = SourceModuleHnNSF(F0_sampled) F0_sampled (batchsize, length, 1) Sine_source (batchsize, length, 1

kokoro/istftnet.py:241

Methodforward

(self, x, s, f0)

kokoro/istftnet.py:299

Methodforward

(self, x)

kokoro/istftnet.py:333

Methodforward

(self, x, s)

kokoro/istftnet.py:378

Methodforward

(self, asr, F0_curve, N, s)

kokoro/istftnet.py:407

Methodforward

Full STFT -> iSTFT pass: returns time-domain reconstruction. Same interface as your original code.

kokoro/custom_stft.py:191

Methodforward

(self, x)

kokoro/modules.py:17

Methodforward

(self, x, input_lengths, m)

kokoro/modules.py:50

Methodforward

(self, x, s)

kokoro/modules.py:79

Methodforward

(self, texts, style, text_lengths, alignment, m)

kokoro/modules.py:109

Methodforward

(self, x, style, text_lengths, m)

kokoro/modules.py:148

Methodforward

(self, *args, **kwargs)

kokoro/modules.py:181

Methodforward

( self, phonemes: str, ref_s: torch.FloatTensor, speed: float = 1, ret

kokoro/model.py:125

Methodforward

( self, input_ids: torch.LongTensor, ref_s: torch.FloatTensor, speed: float =

kokoro/model.py:148

Methodforward

( self, input_values, seq_len, attention_mask=None, mask_time_indices=

src/audio_analysis/wav2vec2.py:13

Methodforward

(self, *args, **kwargs)

src/vram_management/layers.py:58

Methodforward

(self, x, *args, **kwargs)

src/vram_management/layers.py:112

Methodforward

(self, x, *args, **kwargs)

src/vram_management/layers.py:173

Functionfree_model

(model)

wan/distributed/fsdp.py:37

Methodgenerate

r""" Generates video frames from input first-last frame and text prompt using diffusion process. Args: input_prompt (`str

wan/first_last_frame2video.py:133

Methodgenerate

r""" Generates video frames from input image and text prompt using diffusion process. Args: input_prompt (`str`):

wan/image2video.py:133

Methodgenerate

r""" Generates video frames from text prompt using diffusion process. Args: input_prompt (`str`): Text pr

wan/text2video.py:114

Methodgenerate

(self, input_prompt, input_frames, input_masks,

wan/vace.py:773

Methodgenerate_from_tokens

Generate audio from either raw phonemes or pre-processed tokens. Args: tokens: Either a phoneme string or list of pre-pro

kokoro/pipeline.py:247

Functiongenerate_video

(img2vid_image, img2vid_prompt, n_prompt, img2vid_audio_1, img2vid_audio_2, sd_steps, seed

app.py:508

Methodget_current_lora

(self)

wan/wan_lora.py:112

Functionget_mask_from_lengths

(lengths, max_len=None)

src/audio_analysis/torch_utils.py:5

Methodinit_weight

(self, conv)

wan/modules/vae.py:162

Methodinit_weight2

(self, conv)

wan/modules/vae.py:174

Functioninit_weights

(m)

wan/modules/t5.py:32

Functions446 in github.com/MeiGen-AI/MultiTalk