DL-Art-School/codes/models/audio/tts/transformer_diffusion_tts.py

262 lines
12 KiB
Python
Raw Normal View History

2022-05-27 15:49:10 +00:00
import torch
import torch.nn as nn
import torch.nn.functional as F
from models.diffusion.nn import timestep_embedding, normalization, zero_module, conv_nd, linear
from models.diffusion.unet_diffusion import TimestepEmbedSequential, TimestepBlock
from models.lucidrains.x_transformers import Encoder, Attention, FeedForward, RMSScaleShiftNorm, RotaryEmbedding
from trainer.networks import register_model
from utils.util import checkpoint
def is_latent(t):
return t.dtype == torch.float
def is_sequence(t):
return t.dtype == torch.long
class MultiGroupEmbedding(nn.Module):
def __init__(self, tokens, groups, dim):
super().__init__()
self.m = nn.ModuleList([nn.Embedding(tokens, dim // groups) for _ in range(groups)])
def forward(self, x):
h = [embedding(x[:, :, i]) for i, embedding in enumerate(self.m)]
return torch.cat(h, dim=-1)
class TimestepRotaryEmbedSequential(nn.Sequential, TimestepBlock):
def forward(self, x, emb, rotary_emb):
for layer in self:
if isinstance(layer, TimestepBlock):
x = layer(x, emb, rotary_emb)
else:
x = layer(x, rotary_emb)
return x
class AttentionBlock(TimestepBlock):
def __init__(self, dim, heads, dropout):
super().__init__()
self.attn = Attention(dim, heads=heads, causal=False, dropout=dropout, zero_init_output=False)
self.ff = FeedForward(dim, mult=1, dropout=dropout, zero_init_output=True)
self.rms_scale_norm = RMSScaleShiftNorm(dim)
def forward(self, x, timestep_emb, rotary_emb):
h = self.rms_scale_norm(x, norm_scale_shift_inp=timestep_emb)
h, _, _, _ = checkpoint(self.attn, h, None, None, None, None, None, rotary_emb)
h = checkpoint(self.ff, h)
return h + x
class TransformerDiffusion(nn.Module):
"""
A diffusion model composed entirely of stacks of transformer layers. Why would you do it any other way?
"""
def __init__(
self,
model_channels=512,
num_layers=8,
in_channels=256,
in_latent_channels=512,
clvp_in_dim=768,
rotary_emb_dim=32,
token_count=8,
in_groups=None,
2022-05-27 17:12:03 +00:00
types=2,
2022-05-27 15:49:10 +00:00
out_channels=512, # mean and variance
dropout=0,
use_fp16=False,
# Parameters for regularization.
unconditioned_percentage=.1, # This implements a mechanism similar to what is used in classifier-free training.
):
super().__init__()
self.in_channels = in_channels
self.model_channels = model_channels
self.out_channels = out_channels
self.dropout = dropout
self.unconditioned_percentage = unconditioned_percentage
self.enable_fp16 = use_fp16
heads = model_channels//64
self.inp_block = conv_nd(1, in_channels, model_channels, 3, 1, 1)
self.time_embed = nn.Sequential(
linear(model_channels, model_channels),
nn.SiLU(),
linear(model_channels, model_channels),
)
self.conditioning_embedder = nn.Sequential(nn.Conv1d(in_channels, model_channels // 2, 3, padding=1, stride=2),
nn.Conv1d(model_channels//2, model_channels,3,padding=1,stride=2))
self.conditioning_encoder = Encoder(
dim=model_channels,
depth=4,
heads=heads,
ff_dropout=dropout,
attn_dropout=dropout,
use_rmsnorm=True,
ff_glu=True,
rotary_pos_emb=True,
)
self.clvp_encoder = nn.Linear(clvp_in_dim, model_channels)
2022-05-27 17:19:30 +00:00
self.type_embedding = nn.Embedding(types, model_channels)
2022-05-27 15:49:10 +00:00
# Either code_converter or latent_converter is used, depending on what type of conditioning data is fed.
# This model is meant to be able to be trained on both for efficiency purposes - it is far less computationally
# complex to generate tokens, while generating latents will normally mean propagating through a deep autoregressive
# transformer network.
if in_groups is None:
self.embeddings = nn.Embedding(token_count, model_channels)
else:
self.embeddings = MultiGroupEmbedding(token_count, in_groups, model_channels)
self.latent_conditioner = nn.Sequential(
nn.Conv1d(in_latent_channels, model_channels, 3, padding=1),
Encoder(
dim=model_channels,
depth=2,
heads=heads,
ff_dropout=dropout,
attn_dropout=dropout,
use_rmsnorm=True,
ff_glu=True,
rotary_pos_emb=True,
)
)
self.latent_fade = nn.Parameter(torch.zeros(1,1,model_channels))
self.code_converter = Encoder(
dim=model_channels,
depth=3,
heads=heads,
ff_dropout=dropout,
attn_dropout=dropout,
use_rmsnorm=True,
ff_glu=True,
rotary_pos_emb=True,
)
self.unconditioned_embedding = nn.Parameter(torch.randn(1,1,model_channels))
self.mel_head = nn.Conv1d(model_channels, in_channels, kernel_size=3, padding=1)
self.rotary_embeddings = RotaryEmbedding(rotary_emb_dim)
self.intg = nn.Linear(model_channels*2, model_channels)
self.layers = TimestepRotaryEmbedSequential(*[AttentionBlock(model_channels, model_channels//64, dropout) for _ in range(num_layers)])
self.out = nn.Sequential(
normalization(model_channels),
nn.SiLU(),
zero_module(conv_nd(1, model_channels, out_channels, 3, padding=1)),
)
self.debug_codes = {}
def get_grad_norm_parameter_groups(self):
groups = {
'contextual_embedder': list(self.conditioning_embedder.parameters()),
'layers': list(self.layers.parameters()) + list(self.inp_block.parameters()),
'code_converters': list(self.embeddings.parameters()) + list(self.code_converter.parameters()) + list(self.latent_conditioner.parameters()),
'time_embed': list(self.time_embed.parameters()),
}
return groups
def timestep_independent(self, codes, conditioning_input, expected_seq_len, prenet_latent=None, return_code_pred=False):
cond_emb = self.conditioning_embedder(conditioning_input).permute(0,2,1)
cond_emb = self.conditioning_encoder(cond_emb)[:, 0]
code_emb = self.embeddings(codes)
if prenet_latent is not None:
latent_conditioning = self.latent_conditioner(prenet_latent)
code_emb = code_emb + latent_conditioning * self.latent_fade
unconditioned_batches = torch.zeros((code_emb.shape[0], 1, 1), device=code_emb.device)
# Mask out the conditioning branch for whole batch elements, implementing something similar to classifier-free guidance.
if self.training and self.unconditioned_percentage > 0:
unconditioned_batches = torch.rand((code_emb.shape[0], 1, 1),
device=code_emb.device) < self.unconditioned_percentage
code_emb = torch.where(unconditioned_batches, self.unconditioned_embedding.repeat(codes.shape[0], 1, 1),
code_emb)
code_emb = self.code_converter(code_emb)
expanded_code_emb = F.interpolate(code_emb.permute(0,2,1), size=expected_seq_len, mode='nearest').permute(0,2,1)
if not return_code_pred:
return expanded_code_emb, cond_emb
else:
# Perform the mel_head computation on the pre-exanded code embeddings, then interpolate it separately.
mel_pred = self.mel_head(code_emb.permute(0,2,1))
mel_pred = F.interpolate(mel_pred, size=expected_seq_len, mode='nearest')
# Multiply mel_pred by !unconditioned_branches, which drops the gradient on unconditioned branches.
# This is because we don't want that gradient being used to train parameters through the codes_embedder as
# it unbalances contributions to that network from the MSE loss.
mel_pred = mel_pred * unconditioned_batches.logical_not()
return expanded_code_emb, cond_emb, mel_pred
2022-05-27 17:12:03 +00:00
def forward(self, x, timesteps, codes=None, conditioning_input=None, clvp_input=None, type=None, prenet_latent=None, precomputed_code_embeddings=None,
2022-05-27 15:49:10 +00:00
precomputed_cond_embeddings=None, conditioning_free=False, return_code_pred=False):
if precomputed_code_embeddings is not None:
assert precomputed_cond_embeddings is not None, "Must specify both precomputed embeddings if one is specified"
assert codes is None and conditioning_input is None and prenet_latent is None, "Do not provide precomputed embeddings and the other parameters. It is unclear what you want me to do here."
assert not (return_code_pred and precomputed_code_embeddings is not None), "I cannot compute a code_pred output for you."
2022-05-27 17:12:03 +00:00
assert type is not None, "Type is required."
2022-05-27 15:49:10 +00:00
unused_params = []
if not return_code_pred:
unused_params.extend(list(self.mel_head.parameters()))
if conditioning_free:
code_emb = self.unconditioned_embedding.repeat(x.shape[0], 1, x.shape[-1])
unused_params.extend(list(self.code_converter.parameters()) + list(self.code_embedding.parameters()))
unused_params.extend(list(self.latent_conditioner.parameters()))
else:
if precomputed_code_embeddings is not None:
code_emb = precomputed_code_embeddings
cond_emb = precomputed_cond_embeddings
else:
code_emb, cond_emb, mel_pred = self.timestep_independent(codes, conditioning_input, x.shape[-1], prenet_latent, True)
if prenet_latent is None:
unused_params.extend(list(self.latent_conditioner.parameters()) + [self.latent_fade])
unused_params.append(self.unconditioned_embedding)
clvp_emb = torch.zeros_like(cond_emb) if clvp_input is None else self.clvp_encoder(clvp_input)
2022-05-27 17:12:03 +00:00
type_emb = self.type_embedding(type)
2022-05-27 15:49:10 +00:00
if clvp_input is None:
unused_params.extend(self.clvp_encoder.parameters())
2022-05-27 17:12:03 +00:00
blk_emb = self.time_embed(timestep_embedding(timesteps, self.model_channels)) + cond_emb + clvp_emb + type_emb
2022-05-27 15:49:10 +00:00
x = self.inp_block(x).permute(0,2,1)
rotary_pos_emb = self.rotary_embeddings(x.shape[1], x.device)
x = self.intg(torch.cat([x, code_emb], dim=-1))
x = self.layers(x, blk_emb, rotary_pos_emb)
x = x.float().permute(0,2,1)
out = self.out(x)
# Involve probabilistic or possibly unused parameters in loss so we don't get DDP errors.
extraneous_addition = 0
for p in unused_params:
extraneous_addition = extraneous_addition + p.mean()
out = out + extraneous_addition * 0
if return_code_pred:
return out, mel_pred
return out
@register_model
2022-05-27 17:12:03 +00:00
def register_transformer_diffusion_tts(opt_net, opt):
2022-05-27 15:49:10 +00:00
return TransformerDiffusion(**opt_net['kwargs'])
if __name__ == '__main__':
clip = torch.randn(2, 256, 400)
aligned_latent = torch.randn(2,100,512)
aligned_sequence = torch.randint(0,8,(2,100,8))
cond = torch.randn(2, 256, 400)
ts = torch.LongTensor([600, 600])
clvp = torch.randn(2,768)
2022-05-27 17:19:30 +00:00
type = torch.LongTensor([0,1])
2022-05-27 15:49:10 +00:00
model = TransformerDiffusion(512, unconditioned_percentage=.5, in_groups=8)
2022-05-27 17:19:30 +00:00
o = model(clip, ts, aligned_sequence, cond, clvp_input=clvp, type=type, return_code_pred=True)
2022-05-27 15:49:10 +00:00
#o = model(clip, ts, aligned_sequence, cond, aligned_latent)