Functions — github.com/LqNoob/Neural-Codec-and-Speech-Language-Models

hub / github.com/LqNoob/Neural-Codec-and-Speech-Language-Models / functions

Functions116 in github.com/LqNoob/Neural-Codec-and-Speech-Language-Models

⨍ Functions 116◇ Types & classes 29

↓ 12 callersMethod__init__

( self, n_fft: int, hop_length: int, win_length: int, center=True,

generator/vocos/vocos.py:41

↓ 12 callersFunctionget_padding

(kernel_size, dilation=1)

generator/hifigan/utils.py:57

↓ 6 callersMethod__init__

(self, h, channels, kernel_size=3, dilation=(1, 3, 5))

generator/hifigan/hifigan.py:18

↓ 6 callersMethodget_padding

(kernel_size: int, dilation: int = 1)

generator/vocos/vocos.py:700

↓ 3 callersMethoddecode_code

(self, embed_id)

vector_quantizer/factorized_vector_quantize.py:99

↓ 2 callersFunctionWNConv1d

(*args, **kwargs)

vector_quantizer/lookup_free_quantize.py:14

↓ 2 callersFunctionWNConv1d

(*args, **kwargs)

vector_quantizer/vector_quantize.py:14

↓ 2 callersFunctionWNConv1d

(*args, **kwargs)

vector_quantizer/factorized_vector_quantize.py:14

↓ 2 callersMethod__init__

( self, dim, codebook_size, use_l2_normlize=False, )

vector_quantizer/vector_quantize.py:195

↓ 2 callersFunctiondefault

(*args)

vector_quantizer/finite_scalar_quantization.py:25

↓ 2 callersFunctionema_inplace

(moving_avg, new, decay)

vector_quantizer/vector_quantize.py:26

↓ 2 callersMethodinit_embed_

(self, data)

vector_quantizer/vector_quantize.py:109

↓ 2 callersFunctionsample_vectors

(samples, num)

vector_quantizer/vector_quantize.py:34

↓ 2 callersFunctionunpack_one

(t, ps, pattern)

vector_quantizer/finite_scalar_quantization.py:34

↓ 1 callersMethod_scale_and_shift

(self, zhat_normalized: Tensor)

vector_quantizer/finite_scalar_quantization.py:100

↓ 1 callersMethod_scale_and_shift_inverse

(self, zhat: Tensor)

vector_quantizer/finite_scalar_quantization.py:104

↓ 1 callersMethodbound

Bound `z`, an array of shape (..., d).

vector_quantizer/finite_scalar_quantization.py:87

↓ 1 callersMethodcodes_to_indices

Converts a `code` to an index in the codebook.

vector_quantizer/finite_scalar_quantization.py:108

↓ 1 callersMethoddecode_latents

(self, latents)

vector_quantizer/vector_quantize.py:385

↓ 1 callersMethoddecode_latents

(self, latents)

vector_quantizer/factorized_vector_quantize.py:102

↓ 1 callersMethodembed_code

(self, embed_id)

vector_quantizer/factorized_vector_quantize.py:96

↓ 1 callersFunctionexists

(v)

vector_quantizer/finite_scalar_quantization.py:22

↓ 1 callersMethodexpire_codes_

(self, batch_samples)

vector_quantizer/vector_quantize.py:122

↓ 1 callersMethodindices_to_codes

Inverse of `codes_to_indices`.

vector_quantizer/finite_scalar_quantization.py:114

↓ 1 callersFunctionkmeans

(samples, num_clusters, num_iters=10, use_cosine_sim=False)

vector_quantizer/vector_quantize.py:45

↓ 1 callersFunctionl2norm

(t)

vector_quantizer/vector_quantize.py:22

↓ 1 callersFunctionlaplace_smoothing

(x, n_categories, eps=1e-5)

vector_quantizer/vector_quantize.py:30

↓ 1 callersMethodlatent2dist

(self, x)

vector_quantizer/vector_quantize.py:234

↓ 1 callersFunctionpack_one

(t, pattern)

vector_quantizer/finite_scalar_quantization.py:31

↓ 1 callersMethodquantize

Quantizes z, returns quantized zhat, same shape as z.

vector_quantizer/finite_scalar_quantization.py:94

↓ 1 callersMethodremove_weight_norm

(self)

generator/hifigan/hifigan.py:84

↓ 1 callersMethodreplace

(self, samples, mask)

vector_quantizer/vector_quantize.py:116

↓ 1 callersFunctionround_ste

Round with straight through gradients.

vector_quantizer/finite_scalar_quantization.py:39

↓ 1 callersFunctionsymexp

(x: torch.Tensor)

generator/vocos/vocos.py:36

↓ 1 callersMethodvq2emb

(self, vq)

vector_quantizer/vector_quantize.py:230

FunctionWNConvTranspose1d

(*args, **kwargs)

vector_quantizer/lookup_free_quantize.py:18

FunctionWNConvTranspose1d

(*args, **kwargs)

vector_quantizer/vector_quantize.py:18

FunctionWNConvTranspose1d

(*args, **kwargs)

vector_quantizer/factorized_vector_quantize.py:18

Method__init__

( self, input_dim: int = 256, num_quantizers: int = 8, codebook_size: int = 10

vector_quantizer/residual_vq.py:28

Method__init__

( self, input_dim, codebook_size, codebook_dim, )

vector_quantizer/lookup_free_quantize.py:23

Method__init__

( self, levels: List[int], dim: Optional[int] = None, num_codebooks = 1,

vector_quantizer/finite_scalar_quantization.py:47

Method__init__

( self, dim, codebook_size, kmeans_init=False, kmeans_iters=10,

vector_quantizer/vector_quantize.py:77

Method__init__

( self, input_dim, codebook_size, codebook_dim, commitment=0.005,

vector_quantizer/vector_quantize.py:278

Method__init__

( self, input_dim, codebook_size, codebook_dim, commitment=0.005,

vector_quantizer/factorized_vector_quantize.py:23

Method__init__

(self, h, channels, kernel_size=3, dilation=(1, 3))

generator/hifigan/hifigan.py:92

Method__init__

(self, h)

generator/hifigan/hifigan.py:128

Method__init__

(self, period, kernel_size=5, stride=3, use_spectral_norm=False)

generator/hifigan/hifigan.py:192

Method__init__

(self)

generator/hifigan/hifigan.py:245

Method__init__

(self, use_spectral_norm=False)

generator/hifigan/hifigan.py:272

Method__init__

(self)

generator/hifigan/hifigan.py:300

Method__init__

( self, n_fft: int, hop_length: int, win_length: int, padding: str = "same" )

generator/vocos/vocos.py:99

Method__init__

(self, frame_len: int, padding: str = "same")

generator/vocos/vocos.py:179

Method__init__

(self, frame_len: int, padding: str = "same")

generator/vocos/vocos.py:240

Method__init__

(self, dim: int, n_fft: int, hop_length: int, padding: str = "same")

generator/vocos/vocos.py:326

Method__init__

( self, dim: int, mdct_frame_len: int, padding: str = "same", sample_r

generator/vocos/vocos.py:377

Method__init__

( self, dim: int, mdct_frame_len: int, padding: str = "same", clip_aud

generator/vocos/vocos.py:435

Method__init__

( self, dim: int, intermediate_dim: int, layer_scale_init_value: float,

generator/vocos/vocos.py:481

Method__init__

(self, num_embeddings: int, embedding_dim: int, eps: float = 1e-6)

generator/vocos/vocos.py:539

Method__init__

( self, dim: int, kernel_size: int = 3, dilation: Tuple[int, int, int] = (1, 3

generator/vocos/vocos.py:576

Method__init__

( self, input_channels: int, dim: int, intermediate_dim: int, num_laye

generator/vocos/vocos.py:734

Method__init__

( self, input_channels, dim, num_blocks, layer_scale_init_value=None,

generator/vocos/vocos.py:797

Method__init__

( self, input_channels: int = 256, dim: int = 384, intermediate_dim: int = 115

generator/vocos/vocos.py:825

Method_init_weights

(self, m)

generator/vocos/vocos.py:766

Functionapply_weight_norm

(m)

generator/hifigan/utils.py:51

Functiondiscriminator_loss

(disc_real_outputs, disc_generated_outputs)

generator/hifigan/hifigan.py:338

Functionfeature_loss

(fmap_r, fmap_g)

generator/hifigan/hifigan.py:329

Methodforward

Parameters ---------- z : Tensor[B x D x T] n_quantizers : int, optional No. of quantizers to use

vector_quantizer/residual_vq.py:68

Methodforward

(self, z)

vector_quantizer/lookup_free_quantize.py:46

Methodforward

einstein notation b - batch n - sequence (or flattened spatial dimensions) d - feature dimension, which is also log2(

vector_quantizer/finite_scalar_quantization.py:139

Methodforward

(self, x)

vector_quantizer/vector_quantize.py:132

Methodforward

(self, x)

vector_quantizer/vector_quantize.py:209

Methodforward

Parameters ---------- z: torch.Tensor[B x D x T] Returns ------- z_q: torch.Tensor[B x D x T]

vector_quantizer/vector_quantize.py:341

Methodforward

Parameters ---------- z: torch.Tensor[B x D x T] Returns ------- z_q: torch.Tensor[B x D x T]

vector_quantizer/factorized_vector_quantize.py:52

Methodforward

(self, x)

generator/hifigan/hifigan.py:75

Methodforward

(self, x)

generator/hifigan/hifigan.py:115

Methodforward

(self, x)

generator/hifigan/hifigan.py:163

Methodforward

(self, x)

generator/hifigan/hifigan.py:222

Methodforward

(self, y, y_hat)

generator/hifigan/hifigan.py:255

Methodforward

(self, x)

generator/hifigan/hifigan.py:286

Methodforward

(self, y, y_hat)

generator/hifigan/hifigan.py:310

Methodforward

(self, x: torch.Tensor)

generator/vocos/vocos.py:56

Methodforward

Compute the Inverse Short Time Fourier Transform (ISTFT) of a complex spectrogram. Args: spec (Tensor): Input complex sp

generator/vocos/vocos.py:112

Methodforward

Apply the Modified Discrete Cosine Transform (MDCT) to the input audio. Args: audio (Tensor): Input audio waveform of sh

generator/vocos/vocos.py:197

Methodforward

Apply the Inverse Modified Discrete Cosine Transform (IMDCT) to the input MDCT coefficients. Args: X (Tensor): Input MDC

generator/vocos/vocos.py:256

Methodforward

Args: x (Tensor): Input tensor of shape (B, L, H), where B is the batch size, L is the sequence length, a

generator/vocos/vocos.py:302

Methodforward

Forward pass of the ISTFTHead module. Args: x (Tensor): Input tensor of shape (B, L, H), where B is the batch size,

generator/vocos/vocos.py:334

Methodforward

Forward pass of the IMDCTSymExpHead module. Args: x (Tensor): Input tensor of shape (B, L, H), where B is the batch size

generator/vocos/vocos.py:401

Methodforward

Forward pass of the IMDCTCosHead module. Args: x (Tensor): Input tensor of shape (B, L, H), where B is the batch size,

generator/vocos/vocos.py:447

Methodforward

( self, x: torch.Tensor, cond_embedding_id: Optional[torch.Tensor] = None )

generator/vocos/vocos.py:508

Methodforward

(self, x: torch.Tensor, cond_embedding_id: torch.Tensor)

generator/vocos/vocos.py:552

Methodforward

(self, x: torch.Tensor)

generator/vocos/vocos.py:682

Methodforward

Args: x (Tensor): Input tensor of shape (B, C, L), where B is the batch size, C denotes output features,

generator/vocos/vocos.py:707

Methodforward

(self, x: torch.Tensor, **kwargs)

generator/vocos/vocos.py:771

Methodforward

(self, x: torch.Tensor, **kwargs)

generator/vocos/vocos.py:817

Methodforward

(self, x)

generator/vocos/vocos.py:877

Functiongenerator_loss

(disc_outputs)

generator/hifigan/hifigan.py:352

Functioninit_weights

(m, mean=0.0, std=0.01)

generator/hifigan/utils.py:45

Methodlatent2dist

(self, z, n_quantizers=None)

vector_quantizer/residual_vq.py:154

Methodlatent2dist

(self, x)

vector_quantizer/vector_quantize.py:171

Methodlatent2dist

(self, latents)

vector_quantizer/vector_quantize.py:398

next →1–100 of 116, ranked by callers