Integrate new diffusion network

api.py

@@ -49,6 +49,15 @@ def download_models():
         print('Done.')
 
 
+def pad_or_truncate(t, length):
+    if t.shape[-1] == length:
+        return t
+    elif t.shape[-1] < length:
+        return F.pad(t, (0, length-t.shape[-1]))
+    else:
+        return t[..., :length]
+
+
 def load_discrete_vocoder_diffuser(trained_diffusion_steps=4000, desired_diffusion_steps=200, cond_free=True, cond_free_k=1):
     """
     Helper function to load a GaussianDiffusion instance configured for use as a vocoder.
@@ -96,26 +105,25 @@ def fix_autoregressive_output(codes, stop_token):
     return codes
 
 
-def do_spectrogram_diffusion(diffusion_model, diffuser, mel_codes, conditioning_input, temperature=1):
+def do_spectrogram_diffusion(diffusion_model, diffuser, mel_codes, conditioning_samples, temperature=1):
     """
-    Uses the specified diffusion model and DVAE model to convert the provided MEL & conditioning inputs into an audio clip.
+    Uses the specified diffusion model to convert discrete codes into a spectrogram.
     """
     with torch.no_grad():
-        cond_mel = wav_to_univnet_mel(conditioning_input.squeeze(1), do_normalization=False)
+        cond_mels = []
-        # Pad MEL to multiples of 32
+        for sample in conditioning_samples:
-        msl = mel_codes.shape[-1]
+            sample = pad_or_truncate(sample, 102400)
-        dsl = 32
+            cond_mel = wav_to_univnet_mel(sample.to(mel_codes.device), do_normalization=False)
-        gap = dsl - (msl % dsl)
+            cond_mels.append(cond_mel)
-        if gap > 0:
+        cond_mels = torch.stack(cond_mels, dim=1)
-            mel = torch.nn.functional.pad(mel_codes, (0, gap))
 
-        output_shape = (mel.shape[0], 100, mel.shape[-1]*4)
+        output_shape = (mel_codes.shape[0], 100, mel_codes.shape[-1]*4)
-        precomputed_embeddings = diffusion_model.timestep_independent(mel_codes, cond_mel)
+        precomputed_embeddings = diffusion_model.timestep_independent(mel_codes, cond_mels, False)
 
         noise = torch.randn(output_shape, device=mel_codes.device) * temperature
         mel = diffuser.p_sample_loop(diffusion_model, output_shape, noise=noise,
                                       model_kwargs={'precomputed_aligned_embeddings': precomputed_embeddings})
-        return denormalize_tacotron_mel(mel)[:,:,:msl*4]
+        return denormalize_tacotron_mel(mel)[:,:,:mel_codes.shape[-1]*4]
 
 
 class TextToSpeech:
@@ -137,12 +145,9 @@ class TextToSpeech:
                              use_xformers=True).cpu().eval()
         self.clip.load_state_dict(torch.load('.models/clip.pth'))
 
-        self.diffusion = DiffusionTts(model_channels=512, in_channels=100, out_channels=200, in_latent_channels=1024,
+        self.diffusion = DiffusionTts(model_channels=1024, num_layers=10, in_channels=100, out_channels=200,
-                                 channel_mult=[1, 2, 3, 4], num_res_blocks=[3, 3, 3, 3],
+                                      in_latent_channels=1024, in_tokens=8193, dropout=0, use_fp16=False, num_heads=16,
-                                 token_conditioning_resolutions=[1, 4, 8],
+                                      layer_drop=0, unconditioned_percentage=0).cpu().eval()
-                                 dropout=0, attention_resolutions=[4, 8], num_heads=8, kernel_size=3, scale_factor=2,
-                                 time_embed_dim_multiplier=4, unconditioned_percentage=0, conditioning_dim_factor=2,
-                                 conditioning_expansion=1).cpu().eval()
         self.diffusion.load_state_dict(torch.load('.models/diffusion.pth'))
 
         self.vocoder = UnivNetGenerator().cpu()
@@ -164,12 +169,6 @@ class TextToSpeech:
         for vs in voice_samples:
             conds.append(load_conditioning(vs))
         conds = torch.stack(conds, dim=1)
-        cond_diffusion = voice_samples[0].cuda()
-        # The diffusion model expects = 88200 conditioning samples.
-        if cond_diffusion.shape[-1] < 88200:
-            cond_diffusion = F.pad(cond_diffusion, (0, 88200-cond_diffusion.shape[-1]))
-        else:
-            cond_diffusion = cond_diffusion[:, :88200]
 
         diffuser = load_discrete_vocoder_diffuser(desired_diffusion_steps=diffusion_iterations, cond_free=cond_free, cond_free_k=cond_free_k)
 
@@ -211,7 +210,7 @@ class TextToSpeech:
             self.vocoder = self.vocoder.cuda()
             for b in range(best_results.shape[0]):
                 code = best_results[b].unsqueeze(0)
-                mel = do_spectrogram_diffusion(self.diffusion, diffuser, code, cond_diffusion, temperature=diffusion_temperature)
+                mel = do_spectrogram_diffusion(self.diffusion, diffuser, code, voice_samples, temperature=diffusion_temperature)
                 wav = self.vocoder.inference(mel)
                 wav_candidates.append(wav.cpu())
             self.diffusion = self.diffusion.cpu()

models/arch_util.py

@@ -6,6 +6,7 @@ import torch.nn as nn
 import torch.nn.functional as F
 import torchaudio
 from x_transformers import ContinuousTransformerWrapper
+from x_transformers.x_transformers import RelativePositionBias
 
 
 def zero_module(module):
@@ -49,7 +50,7 @@ class QKVAttentionLegacy(nn.Module):
         super().__init__()
         self.n_heads = n_heads
 
-    def forward(self, qkv, mask=None):
+    def forward(self, qkv, mask=None, rel_pos=None):
         """
         Apply QKV attention.
 
@@ -64,6 +65,8 @@ class QKVAttentionLegacy(nn.Module):
         weight = torch.einsum(
             "bct,bcs->bts", q * scale, k * scale
         )  # More stable with f16 than dividing afterwards
+        if rel_pos is not None:
+            weight = rel_pos(weight.reshape(bs, self.n_heads, weight.shape[-2], weight.shape[-1])).reshape(bs * self.n_heads, weight.shape[-2], weight.shape[-1])
         weight = torch.softmax(weight.float(), dim=-1).type(weight.dtype)
         if mask is not None:
             # The proper way to do this is to mask before the softmax using -inf, but that doesn't work properly on CPUs.
@@ -87,9 +90,12 @@ class AttentionBlock(nn.Module):
         channels,
         num_heads=1,
         num_head_channels=-1,
+        do_checkpoint=True,
+        relative_pos_embeddings=False,
     ):
         super().__init__()
         self.channels = channels
+        self.do_checkpoint = do_checkpoint
         if num_head_channels == -1:
             self.num_heads = num_heads
         else:
@@ -99,21 +105,20 @@ class AttentionBlock(nn.Module):
             self.num_heads = channels // num_head_channels
         self.norm = normalization(channels)
         self.qkv = nn.Conv1d(channels, channels * 3, 1)
+        # split heads before split qkv
         self.attention = QKVAttentionLegacy(self.num_heads)
 
         self.proj_out = zero_module(nn.Conv1d(channels, channels, 1))
+        if relative_pos_embeddings:
+            self.relative_pos_embeddings = RelativePositionBias(scale=(channels // self.num_heads) ** .5, causal=False, heads=num_heads, num_buckets=32, max_distance=64)
+        else:
+            self.relative_pos_embeddings = None
 
     def forward(self, x, mask=None):
-        if mask is not None:
-            return self._forward(x, mask)
-        else:
-            return self._forward(x)
-
-    def _forward(self, x, mask=None):
         b, c, *spatial = x.shape
         x = x.reshape(b, c, -1)
         qkv = self.qkv(self.norm(x))
-        h = self.attention(qkv, mask)
+        h = self.attention(qkv, mask, self.relative_pos_embeddings)
         h = self.proj_out(h)
         return (x + h).reshape(b, c, *spatial)
 

models/diffusion_decoder.py

@@ -1,22 +1,13 @@
-"""
-This model is based on OpenAI's UNet from improved diffusion, with modifications to support a MEL conditioning signal
-and an audio conditioning input. It has also been simplified somewhat.
-Credit: https://github.com/openai/improved-diffusion
-"""
-import functools
 import math
+import random
 from abc import abstractmethod
 
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
 from torch import autocast
-from torch.nn import Linear
-from torch.utils.checkpoint import checkpoint
-from x_transformers import ContinuousTransformerWrapper, Encoder
 
-from models.arch_util import normalization, zero_module, Downsample, Upsample, AudioMiniEncoder, AttentionBlock, \
+from models.arch_util import normalization, AttentionBlock
-    CheckpointedXTransformerEncoder
 
 
 def is_latent(t):
@@ -27,13 +18,6 @@ def is_sequence(t):
     return t.dtype == torch.long
 
 
-def ceil_multiple(base, multiple):
-    res = base % multiple
-    if res == 0:
-        return base
-    return base + (multiple - res)
-
-
 def timestep_embedding(timesteps, dim, max_period=10000):
     """
     Create sinusoidal timestep embeddings.
@@ -56,10 +40,6 @@ def timestep_embedding(timesteps, dim, max_period=10000):
 
 
 class TimestepBlock(nn.Module):
-    """
-    Any module where forward() takes timestep embeddings as a second argument.
-    """
-
     @abstractmethod
     def forward(self, x, emb):
         """
@@ -68,11 +48,6 @@ class TimestepBlock(nn.Module):
 
 
 class TimestepEmbedSequential(nn.Sequential, TimestepBlock):
-    """
-    A sequential module that passes timestep embeddings to the children that
-    support it as an extra input.
-    """
-
     def forward(self, x, emb):
         for layer in self:
             if isinstance(layer, TimestepBlock):
@@ -89,6 +64,7 @@ class ResBlock(TimestepBlock):
         emb_channels,
         dropout,
         out_channels=None,
+        dims=2,
         kernel_size=3,
         efficient_config=True,
         use_scale_shift_norm=False,
@@ -111,7 +87,7 @@ class ResBlock(TimestepBlock):
 
         self.emb_layers = nn.Sequential(
             nn.SiLU(),
-            Linear(
+            nn.Linear(
                 emb_channels,
                 2 * self.out_channels if use_scale_shift_norm else self.out_channels,
             ),
@@ -120,9 +96,7 @@ class ResBlock(TimestepBlock):
             normalization(self.out_channels),
             nn.SiLU(),
             nn.Dropout(p=dropout),
-            zero_module(
+                nn.Conv1d(self.out_channels, self.out_channels, kernel_size, padding=padding),
-                nn.Conv1d(self.out_channels, self.out_channels, kernel_size, padding=padding)
-            ),
         )
 
         if self.out_channels == channels:
@@ -131,18 +105,6 @@ class ResBlock(TimestepBlock):
             self.skip_connection = nn.Conv1d(channels, self.out_channels, eff_kernel, padding=eff_padding)
 
     def forward(self, x, emb):
-        """
-        Apply the block to a Tensor, conditioned on a timestep embedding.
-
-        :param x: an [N x C x ...] Tensor of features.
-        :param emb: an [N x emb_channels] Tensor of timestep embeddings.
-        :return: an [N x C x ...] Tensor of outputs.
-        """
-        return checkpoint(
-            self._forward, x, emb
-        )
-
-    def _forward(self, x, emb):
         h = self.in_layers(x)
         emb_out = self.emb_layers(emb).type(h.dtype)
         while len(emb_out.shape) < len(h.shape):
@@ -158,372 +120,205 @@ class ResBlock(TimestepBlock):
         return self.skip_connection(x) + h
 
 
+class DiffusionLayer(TimestepBlock):
+    def __init__(self, model_channels, dropout, num_heads):
+        super().__init__()
+        self.resblk = ResBlock(model_channels, model_channels, dropout, model_channels, dims=1, use_scale_shift_norm=True)
+        self.attn = AttentionBlock(model_channels, num_heads, relative_pos_embeddings=True)
+
+    def forward(self, x, time_emb):
+        y = self.resblk(x, time_emb)
+        return self.attn(y)
+
+
 class DiffusionTts(nn.Module):
-    """
-    The full UNet model with attention and timestep embedding.
-
-    Customized to be conditioned on an aligned prior derived from a autoregressive
-    GPT-style model.
-
-    :param in_channels: channels in the input Tensor.
-    :param in_latent_channels: channels from the input latent.
-    :param model_channels: base channel count for the model.
-    :param out_channels: channels in the output Tensor.
-    :param num_res_blocks: number of residual blocks per downsample.
-    :param attention_resolutions: a collection of downsample rates at which
-        attention will take place. May be a set, list, or tuple.
-        For example, if this contains 4, then at 4x downsampling, attention
-        will be used.
-    :param dropout: the dropout probability.
-    :param channel_mult: channel multiplier for each level of the UNet.
-    :param conv_resample: if True, use learned convolutions for upsampling and
-        downsampling.
-    :param num_heads: the number of attention heads in each attention layer.
-    :param num_heads_channels: if specified, ignore num_heads and instead use
-                               a fixed channel width per attention head.
-    :param num_heads_upsample: works with num_heads to set a different number
-                               of heads for upsampling. Deprecated.
-    :param use_scale_shift_norm: use a FiLM-like conditioning mechanism.
-    :param resblock_updown: use residual blocks for up/downsampling.
-    :param use_new_attention_order: use a different attention pattern for potentially
-                                    increased efficiency.
-    """
-
     def __init__(
             self,
-            model_channels,
+            model_channels=512,
-            in_channels=1,
+            num_layers=8,
-            in_latent_channels=1024,
+            in_channels=100,
+            in_latent_channels=512,
             in_tokens=8193,
-            conditioning_dim_factor=8,
+            out_channels=200,  # mean and variance
-            conditioning_expansion=4,
-            out_channels=2,  # mean and variance
             dropout=0,
-            # res           1, 2, 4, 8,16,32,64,128,256,512, 1K, 2K
-            channel_mult=  (1,1.5,2, 3, 4, 6, 8, 12, 16, 24, 32, 48),
-            num_res_blocks=(1, 1, 1, 1, 1, 2, 2, 2,   2,  2,  2,  2),
-            # spec_cond:    1, 0, 0, 1, 0, 0, 1, 0,   0,  1,  0,  0)
-            # attn:         0, 0, 0, 0, 0, 0, 0, 0,   0,  1,  1,  1
-            token_conditioning_resolutions=(1,16,),
-            attention_resolutions=(512,1024,2048),
-            conv_resample=True,
             use_fp16=False,
-            num_heads=1,
+            num_heads=16,
-            num_head_channels=-1,
-            num_heads_upsample=-1,
-            kernel_size=3,
-            scale_factor=2,
-            time_embed_dim_multiplier=4,
-            freeze_main_net=False,
-            efficient_convs=True,  # Uses kernels with width of 1 in several places rather than 3.
-            use_scale_shift_norm=True,
             # Parameters for regularization.
+            layer_drop=.1,
             unconditioned_percentage=.1,  # This implements a mechanism similar to what is used in classifier-free training.
-            # Parameters for super-sampling.
-            super_sampling=False,
-            super_sampling_max_noising_factor=.1,
     ):
         super().__init__()
 
-        if num_heads_upsample == -1:
-            num_heads_upsample = num_heads
-
-        if super_sampling:
-            in_channels *= 2  # In super-sampling mode, the LR input is concatenated directly onto the input.
         self.in_channels = in_channels
         self.model_channels = model_channels
         self.out_channels = out_channels
-        self.attention_resolutions = attention_resolutions
         self.dropout = dropout
-        self.channel_mult = channel_mult
-        self.conv_resample = conv_resample
         self.num_heads = num_heads
-        self.num_head_channels = num_head_channels
-        self.num_heads_upsample = num_heads_upsample
-        self.super_sampling_enabled = super_sampling
-        self.super_sampling_max_noising_factor = super_sampling_max_noising_factor
         self.unconditioned_percentage = unconditioned_percentage
         self.enable_fp16 = use_fp16
-        self.alignment_size = 2 ** (len(channel_mult)+1)
+        self.layer_drop = layer_drop
-        self.freeze_main_net = freeze_main_net
-        padding = 1 if kernel_size == 3 else 2
-        down_kernel = 1 if efficient_convs else 3
 
-        time_embed_dim = model_channels * time_embed_dim_multiplier
+        self.inp_block = nn.Conv1d(in_channels, model_channels, 3, 1, 1)
         self.time_embed = nn.Sequential(
-            Linear(model_channels, time_embed_dim),
+            nn.Linear(model_channels, model_channels),
             nn.SiLU(),
-            Linear(time_embed_dim, time_embed_dim),
+            nn.Linear(model_channels, model_channels),
         )
 
-        conditioning_dim = model_channels * conditioning_dim_factor
         # 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.
+        self.code_embedding = nn.Embedding(in_tokens, model_channels)
         self.code_converter = nn.Sequential(
-            nn.Embedding(in_tokens, conditioning_dim),
+            AttentionBlock(model_channels, num_heads, relative_pos_embeddings=True),
-            CheckpointedXTransformerEncoder(
+            AttentionBlock(model_channels, num_heads, relative_pos_embeddings=True),
-                needs_permute=False,
+            AttentionBlock(model_channels, num_heads, relative_pos_embeddings=True),
-                max_seq_len=-1,
+        )
-                use_pos_emb=False,
+        self.code_norm = normalization(model_channels)
-                attn_layers=Encoder(
+        self.latent_converter = nn.Conv1d(in_latent_channels, model_channels, 1)
-                    dim=conditioning_dim,
+        self.contextual_embedder = nn.Sequential(nn.Conv1d(in_channels,model_channels,3,padding=1,stride=2),
-                    depth=3,
+                                                 nn.Conv1d(model_channels, model_channels*2,3,padding=1,stride=2),
-                    heads=num_heads,
+                                                 AttentionBlock(model_channels*2, num_heads, relative_pos_embeddings=True, do_checkpoint=False),
-                    ff_dropout=dropout,
+                                                 AttentionBlock(model_channels*2, num_heads, relative_pos_embeddings=True, do_checkpoint=False),
-                    attn_dropout=dropout,
+                                                 AttentionBlock(model_channels*2, num_heads, relative_pos_embeddings=True, do_checkpoint=False),
-                    use_rmsnorm=True,
+                                                 AttentionBlock(model_channels*2, num_heads, relative_pos_embeddings=True, do_checkpoint=False),
-                    ff_glu=True,
+                                                 AttentionBlock(model_channels*2, num_heads, relative_pos_embeddings=True, do_checkpoint=False))
-                    rotary_emb_dim=True,
+        self.unconditioned_embedding = nn.Parameter(torch.randn(1,model_channels,1))
-                )
-            ))
-        self.latent_converter = nn.Conv1d(in_latent_channels, conditioning_dim, 1)
-        self.aligned_latent_padding_embedding = nn.Parameter(torch.randn(1,in_latent_channels,1))
-        if in_channels > 60:  # It's a spectrogram.
-            self.contextual_embedder = nn.Sequential(nn.Conv1d(in_channels,conditioning_dim,3,padding=1,stride=2),
-                                                     CheckpointedXTransformerEncoder(
-                                                         needs_permute=True,
-                                                         max_seq_len=-1,
-                                                         use_pos_emb=False,
-                                                         attn_layers=Encoder(
-                                                             dim=conditioning_dim,
-                                                             depth=4,
-                                                             heads=num_heads,
-                                                             ff_dropout=dropout,
-                                                             attn_dropout=dropout,
-                                                             use_rmsnorm=True,
-                                                             ff_glu=True,
-                                                             rotary_emb_dim=True,
-                                                         )
-                                                     ))
-        else:
-            self.contextual_embedder = AudioMiniEncoder(1, conditioning_dim, base_channels=32, depth=6, resnet_blocks=1,
-                                                        attn_blocks=3, num_attn_heads=8, dropout=dropout, downsample_factor=4, kernel_size=5)
-        self.conditioning_conv = nn.Conv1d(conditioning_dim*2, conditioning_dim, 1)
-        self.unconditioned_embedding = nn.Parameter(torch.randn(1,conditioning_dim,1))
         self.conditioning_timestep_integrator = TimestepEmbedSequential(
-                    ResBlock(conditioning_dim, time_embed_dim, dropout, out_channels=conditioning_dim, kernel_size=1, use_scale_shift_norm=use_scale_shift_norm),
+            DiffusionLayer(model_channels, dropout, num_heads),
-                    AttentionBlock(conditioning_dim, num_heads=num_heads, num_head_channels=num_head_channels),
+            DiffusionLayer(model_channels, dropout, num_heads),
-                    ResBlock(conditioning_dim, time_embed_dim, dropout, out_channels=conditioning_dim, kernel_size=1, use_scale_shift_norm=use_scale_shift_norm),
+            DiffusionLayer(model_channels, dropout, num_heads),
-                    AttentionBlock(conditioning_dim, num_heads=num_heads, num_head_channels=num_head_channels),
-                    ResBlock(conditioning_dim, time_embed_dim, dropout, out_channels=conditioning_dim, kernel_size=1, use_scale_shift_norm=use_scale_shift_norm),
         )
-        self.conditioning_expansion = conditioning_expansion
+        self.integrating_conv = nn.Conv1d(model_channels*2, model_channels, kernel_size=1)
+        self.mel_head = nn.Conv1d(model_channels, in_channels, kernel_size=3, padding=1)
 
-        self.input_blocks = nn.ModuleList(
+        self.layers = nn.ModuleList([DiffusionLayer(model_channels, dropout, num_heads) for _ in range(num_layers)] +
-            [
+                                    [ResBlock(model_channels, model_channels, dropout, dims=1, use_scale_shift_norm=True) for _ in range(3)])
-                TimestepEmbedSequential(
-                    nn.Conv1d(in_channels, model_channels, kernel_size, padding=padding)
-                )
-            ]
-        )
-        token_conditioning_blocks = []
-        self._feature_size = model_channels
-        input_block_chans = [model_channels]
-        ch = model_channels
-        ds = 1
-
-        for level, (mult, num_blocks) in enumerate(zip(channel_mult, num_res_blocks)):
-            if ds in token_conditioning_resolutions:
-                token_conditioning_block = nn.Conv1d(conditioning_dim, ch, 1)
-                token_conditioning_block.weight.data *= .02
-                self.input_blocks.append(token_conditioning_block)
-                token_conditioning_blocks.append(token_conditioning_block)
-
-            for _ in range(num_blocks):
-                layers = [
-                    ResBlock(
-                        ch,
-                        time_embed_dim,
-                        dropout,
-                        out_channels=int(mult * model_channels),
-                        kernel_size=kernel_size,
-                        efficient_config=efficient_convs,
-                        use_scale_shift_norm=use_scale_shift_norm,
-                    )
-                ]
-                ch = int(mult * model_channels)
-                if ds in attention_resolutions:
-                    layers.append(
-                        AttentionBlock(
-                            ch,
-                            num_heads=num_heads,
-                            num_head_channels=num_head_channels,
-                        )
-                    )
-                self.input_blocks.append(TimestepEmbedSequential(*layers))
-                self._feature_size += ch
-                input_block_chans.append(ch)
-            if level != len(channel_mult) - 1:
-                out_ch = ch
-                self.input_blocks.append(
-                    TimestepEmbedSequential(
-                        Downsample(
-                            ch, conv_resample, out_channels=out_ch, factor=scale_factor, ksize=down_kernel, pad=0 if down_kernel == 1 else 1
-                        )
-                    )
-                )
-                ch = out_ch
-                input_block_chans.append(ch)
-                ds *= 2
-                self._feature_size += ch
-
-        self.middle_block = TimestepEmbedSequential(
-            ResBlock(
-                ch,
-                time_embed_dim,
-                dropout,
-                kernel_size=kernel_size,
-                efficient_config=efficient_convs,
-                use_scale_shift_norm=use_scale_shift_norm,
-            ),
-            AttentionBlock(
-                ch,
-                num_heads=num_heads,
-                num_head_channels=num_head_channels,
-            ),
-            ResBlock(
-                ch,
-                time_embed_dim,
-                dropout,
-                kernel_size=kernel_size,
-                efficient_config=efficient_convs,
-                use_scale_shift_norm=use_scale_shift_norm,
-            ),
-        )
-        self._feature_size += ch
-
-        self.output_blocks = nn.ModuleList([])
-        for level, (mult, num_blocks) in list(enumerate(zip(channel_mult, num_res_blocks)))[::-1]:
-            for i in range(num_blocks + 1):
-                ich = input_block_chans.pop()
-                layers = [
-                    ResBlock(
-                        ch + ich,
-                        time_embed_dim,
-                        dropout,
-                        out_channels=int(model_channels * mult),
-                        kernel_size=kernel_size,
-                        efficient_config=efficient_convs,
-                        use_scale_shift_norm=use_scale_shift_norm,
-                    )
-                ]
-                ch = int(model_channels * mult)
-                if ds in attention_resolutions:
-                    layers.append(
-                        AttentionBlock(
-                            ch,
-                            num_heads=num_heads_upsample,
-                            num_head_channels=num_head_channels,
-                        )
-                    )
-                if level and i == num_blocks:
-                    out_ch = ch
-                    layers.append(
-                        Upsample(ch, conv_resample, out_channels=out_ch, factor=scale_factor)
-                    )
-                    ds //= 2
-                self.output_blocks.append(TimestepEmbedSequential(*layers))
-                self._feature_size += ch
 
         self.out = nn.Sequential(
-            normalization(ch),
+            normalization(model_channels),
             nn.SiLU(),
-            zero_module(nn.Conv1d(model_channels, out_channels, kernel_size, padding=padding)),
+            nn.Conv1d(model_channels, out_channels, 3, padding=1),
         )
 
-    def fix_alignment(self, x, aligned_conditioning):
+    def get_grad_norm_parameter_groups(self):
-        """
+        groups = {
-        The UNet requires that the input <x> is a certain multiple of 2, defined by the UNet depth. Enforce this by
+            'minicoder': list(self.contextual_embedder.parameters()),
-        padding both <x> and <aligned_conditioning> before forward propagation and removing the padding before returning.
+            'layers': list(self.layers.parameters()),
-        """
+            'code_converters': list(self.code_embedding.parameters()) + list(self.code_converter.parameters()) + list(self.latent_converter.parameters()) + list(self.latent_converter.parameters()),
-        cm = ceil_multiple(x.shape[-1], self.alignment_size)
+            'timestep_integrator': list(self.conditioning_timestep_integrator.parameters()) + list(self.integrating_conv.parameters()),
-        if cm != 0:
+            'time_embed': list(self.time_embed.parameters()),
-            pc = (cm-x.shape[-1])/x.shape[-1]
+        }
-            x = F.pad(x, (0,cm-x.shape[-1]))
+        return groups
-            # Also fix aligned_latent, which is aligned to x.
-            if is_latent(aligned_conditioning):
-                aligned_conditioning = torch.cat([aligned_conditioning,
-                                                  self.aligned_latent_padding_embedding.repeat(x.shape[0], 1, int(pc * aligned_conditioning.shape[-1]))], dim=-1)
-            else:
-                aligned_conditioning = F.pad(aligned_conditioning, (0, int(pc*aligned_conditioning.shape[-1])))
-        return x, aligned_conditioning
 
-    def timestep_independent(self, aligned_conditioning, conditioning_input):
+    def timestep_independent(self, aligned_conditioning, conditioning_input, return_code_pred):
         # Shuffle aligned_latent to BxCxS format
         if is_latent(aligned_conditioning):
             aligned_conditioning = aligned_conditioning.permute(0, 2, 1)
 
-        with autocast(aligned_conditioning.device.type, enabled=self.enable_fp16):
+        # Note: this block does not need to repeated on inference, since it is not timestep-dependent or x-dependent.
-            cond_emb = self.contextual_embedder(conditioning_input)
+        speech_conditioning_input = conditioning_input.unsqueeze(1) if len(
-            if len(cond_emb.shape) == 3:  # Just take the first element.
+            conditioning_input.shape) == 3 else conditioning_input
-                cond_emb = cond_emb[:, :, 0]
+        conds = []
-            if is_latent(aligned_conditioning):
+        for j in range(speech_conditioning_input.shape[1]):
-                code_emb = self.latent_converter(aligned_conditioning)
+            conds.append(self.contextual_embedder(speech_conditioning_input[:, j]))
-            else:
+        conds = torch.cat(conds, dim=-1)
-                code_emb = self.code_converter(aligned_conditioning)
+        cond_emb = conds.mean(dim=-1)
-            cond_emb = cond_emb.unsqueeze(-1).repeat(1, 1, code_emb.shape[-1])
+        cond_scale, cond_shift = torch.chunk(cond_emb, 2, dim=1)
-            code_emb = self.conditioning_conv(torch.cat([cond_emb, code_emb], dim=1))
+        if is_latent(aligned_conditioning):
-            return code_emb
+            code_emb = self.latent_converter(aligned_conditioning)
+        else:
+            code_emb = self.code_embedding(aligned_conditioning).permute(0, 2, 1)
+            code_emb = self.code_converter(code_emb)
+        code_emb = self.code_norm(code_emb) * (1 + cond_scale.unsqueeze(-1)) + cond_shift.unsqueeze(-1)
 
-    def forward(self, x, timesteps, precomputed_aligned_embeddings, conditioning_free=False):
+        unconditioned_batches = torch.zeros((code_emb.shape[0], 1, 1), device=code_emb.device)
-        assert x.shape[-1] % self.alignment_size == 0
+        # 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(aligned_conditioning.shape[0], 1, 1),
+                                   code_emb)
+        expanded_code_emb = F.interpolate(code_emb, size=aligned_conditioning.shape[-1]*4, mode='nearest')
 
-        with autocast(x.device.type, enabled=self.enable_fp16):
+        if not return_code_pred:
-            if conditioning_free:
+            return expanded_code_emb
-                code_emb = self.unconditioned_embedding.repeat(x.shape[0], 1, 1)
+        else:
-            else:
+            mel_pred = self.mel_head(expanded_code_emb)
+            # 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, mel_pred
+
+
+    def forward(self, x, timesteps, aligned_conditioning=None, conditioning_input=None, precomputed_aligned_embeddings=None, conditioning_free=False, return_code_pred=False):
+        """
+        Apply the model to an input batch.
+
+        :param x: an [N x C x ...] Tensor of inputs.
+        :param timesteps: a 1-D batch of timesteps.
+        :param aligned_conditioning: an aligned latent or sequence of tokens providing useful data about the sample to be produced.
+        :param conditioning_input: a full-resolution audio clip that is used as a reference to the style you want decoded.
+        :param precomputed_aligned_embeddings: Embeddings returned from self.timestep_independent()
+        :param conditioning_free: When set, all conditioning inputs (including tokens and conditioning_input) will not be considered.
+        :return: an [N x C x ...] Tensor of outputs.
+        """
+        assert precomputed_aligned_embeddings is not None or (aligned_conditioning is not None and conditioning_input is not None)
+        assert not (return_code_pred and precomputed_aligned_embeddings is not None)  # These two are mutually exclusive.
+
+        unused_params = []
+        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_converter.parameters()))
+        else:
+            if precomputed_aligned_embeddings is not None:
                 code_emb = precomputed_aligned_embeddings
-
+            else:
-            time_emb = self.time_embed(timestep_embedding(timesteps, self.model_channels))
+                code_emb, mel_pred = self.timestep_independent(aligned_conditioning, conditioning_input, True)
-            code_emb = torch.repeat_interleave(code_emb, self.conditioning_expansion, dim=-1)
+                if is_latent(aligned_conditioning):
-            code_emb = self.conditioning_timestep_integrator(code_emb, time_emb)
+                    unused_params.extend(list(self.code_converter.parameters()) + list(self.code_embedding.parameters()))
-
-            first = True
-            time_emb = time_emb.float()
-            h = x
-            hs = []
-            for k, module in enumerate(self.input_blocks):
-                if isinstance(module, nn.Conv1d):
-                    h_tok = F.interpolate(module(code_emb), size=(h.shape[-1]), mode='nearest')
-                    h = h + h_tok
                 else:
-                    with autocast(x.device.type, enabled=self.enable_fp16 and not first):
+                    unused_params.extend(list(self.latent_converter.parameters()))
-                        # First block has autocast disabled to allow a high precision signal to be properly vectorized.
+            unused_params.append(self.unconditioned_embedding)
-                        h = module(h, time_emb)
-                    hs.append(h)
-                first = False
-            h = self.middle_block(h, time_emb)
-            for module in self.output_blocks:
-                h = torch.cat([h, hs.pop()], dim=1)
-                h = module(h, time_emb)
 
-        # Last block also has autocast disabled for high-precision outputs.
+        time_emb = self.time_embed(timestep_embedding(timesteps, self.model_channels))
-        h = h.float()
+        code_emb = self.conditioning_timestep_integrator(code_emb, time_emb)
-        out = self.out(h)
+        x = self.inp_block(x)
+        x = torch.cat([x, code_emb], dim=1)
+        x = self.integrating_conv(x)
+        for i, lyr in enumerate(self.layers):
+            # Do layer drop where applicable. Do not drop first and last layers.
+            if self.training and self.layer_drop > 0 and i != 0 and i != (len(self.layers)-1) and random.random() < self.layer_drop:
+                unused_params.extend(list(lyr.parameters()))
+            else:
+                # First and last blocks will have autocast disabled for improved precision.
+                with autocast(x.device.type, enabled=self.enable_fp16 and i != 0):
+                    x = lyr(x, time_emb)
 
+        x = x.float()
+        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
 
 
 if __name__ == '__main__':
-    clip = torch.randn(2, 1, 32868)
+    clip = torch.randn(2, 100, 400)
-    aligned_latent = torch.randn(2,388,1024)
+    aligned_latent = torch.randn(2,388,512)
-    aligned_sequence = torch.randint(0,8192,(2,388))
+    aligned_sequence = torch.randint(0,8192,(2,100))
-    cond = torch.randn(2, 1, 44000)
+    cond = torch.randn(2, 100, 400)
     ts = torch.LongTensor([600, 600])
-    model = DiffusionTts(128,
+    model = DiffusionTts(512, layer_drop=.3, unconditioned_percentage=.5)
-                         channel_mult=[1,1.5,2, 3, 4, 6, 8],
-                         num_res_blocks=[2, 2, 2, 2, 2, 2, 1],
-                         token_conditioning_resolutions=[1,4,16,64],
-                         attention_resolutions=[],
-                         num_heads=8,
-                         kernel_size=3,
-                         scale_factor=2,
-                         time_embed_dim_multiplier=4,
-                         super_sampling=False,
-                         efficient_convs=False)
     # Test with latent aligned conditioning
-    o = model(clip, ts, aligned_latent, cond)
+    #o = model(clip, ts, aligned_latent, cond)
     # Test with sequence aligned conditioning
     o = model(clip, ts, aligned_sequence, cond)
+