From a1d008808742906885c8f35f29d5d8120aedc59d Mon Sep 17 00:00:00 2001 From: Edresson Casanova Date: Mon, 6 Jun 2022 15:10:00 -0300 Subject: [PATCH] Remove VITS End2End loss --- TTS/tts/layers/losses.py | 40 +------------------- TTS/tts/models/vits.py | 80 +++------------------------------------- 2 files changed, 6 insertions(+), 114 deletions(-) diff --git a/TTS/tts/layers/losses.py b/TTS/tts/layers/losses.py index 89a11139..0d8fbe0b 100644 --- a/TTS/tts/layers/losses.py +++ b/TTS/tts/layers/losses.py @@ -685,7 +685,6 @@ class VitsGeneratorLoss(nn.Module): scores_disc_mp=None, feats_disc_mp=None, feats_disc_zp=None, - end2end_info=None, ): """ Shapes: @@ -762,32 +761,6 @@ class VitsGeneratorLoss(nn.Module): loss += kl_vae_loss return_dict["loss_kl_vae"] = kl_vae_loss - if end2end_info is not None: - - # gen loss - loss_gen_end2end = self.generator_loss(scores_fake=end2end_info["scores_disc_fake"])[0] * self.gen_loss_alpha - return_dict["loss_gen_end2end"] = loss_gen_end2end - loss += loss_gen_end2end - - # if do not uses soft dtw - if end2end_info["z_predicted"] is not None: - # loss KL using GT durations - z = end2end_info["z"].float() - logs_q = end2end_info["logs_q"].float() - z_predicted = end2end_info["z_predicted"].float() - logs_p = end2end_info["logs_p"].float() - z_mask = end2end_info["z_mask"].float() - - kl = logs_p - logs_q - 0.5 - kl += 0.5 * ((z - z_predicted) ** 2) * torch.exp(-2.0 * logs_p) - kl = torch.sum(kl * z_mask) - loss_kl_end2end_gt_durations = kl / torch.sum(z_mask) - return_dict["loss_kl_end2end_gt_durations"] = loss_kl_end2end_gt_durations - loss += loss_kl_end2end_gt_durations - else: - pass - # ToDo: implement soft dtw - # pass losses to the dict return_dict["loss_gen"] = loss_gen return_dict["loss_kl"] = loss_kl @@ -822,7 +795,7 @@ class VitsDiscriminatorLoss(nn.Module): fake_losses.append(fake_loss.item()) return loss, real_losses, fake_losses - def forward(self, scores_disc_real, scores_disc_fake, scores_disc_zp=None, scores_disc_mp=None, end2end_info=None): + def forward(self, scores_disc_real, scores_disc_fake, scores_disc_zp=None, scores_disc_mp=None): return_dict = {} return_dict["loss"] = 0.0 loss_disc, loss_disc_real, _ = self.discriminator_loss( @@ -844,17 +817,6 @@ class VitsDiscriminatorLoss(nn.Module): return_dict["loss_disc_latent"] = loss_disc_latent * self.disc_latent_loss_alpha return_dict["loss"] += return_dict["loss_disc_latent"] - if end2end_info is not None: - loss_disc_end2end, loss_disc_real_end2end, _ = self.discriminator_loss( - scores_real=end2end_info["scores_disc_real"], scores_fake=end2end_info["scores_disc_fake"], - ) - return_dict["loss_disc_end2end"] = loss_disc_end2end * self.disc_loss_alpha - return_dict["loss"] += return_dict["loss_disc_end2end"] - - for i, ldr in enumerate(loss_disc_real_end2end): - return_dict[f"loss_disc_end2end_real_{i}"] = ldr - - return return_dict diff --git a/TTS/tts/models/vits.py b/TTS/tts/models/vits.py index ca5d607e..526f8e7b 100644 --- a/TTS/tts/models/vits.py +++ b/TTS/tts/models/vits.py @@ -562,10 +562,6 @@ class VitsArgs(Coqpit): use_prosody_conditional_flow_module: bool = False prosody_conditional_flow_module_on_decoder: bool = False - # end 2 end loss - use_end2end_loss: bool = False - use_soft_dtw: bool = False - use_latent_discriminator: bool = False detach_dp_input: bool = True @@ -1069,7 +1065,6 @@ class Vits(BaseTTS): return g def forward_mas(self, outputs, z_p, m_p, logs_p, x, x_mask, y_mask, g, lang_emb): - predicted_durations = None # find the alignment path attn_mask = torch.unsqueeze(x_mask, -1) * torch.unsqueeze(y_mask, 2) with torch.no_grad(): @@ -1092,15 +1087,6 @@ class Vits(BaseTTS): lang_emb=lang_emb.detach() if self.args.detach_dp_input and lang_emb is not None else lang_emb, ) loss_duration = loss_duration / torch.sum(x_mask) - if self.args.use_end2end_loss: - predicted_durations = self.duration_predictor( - x.detach() if self.args.detach_dp_input else x, - x_mask, - g=g.detach() if self.args.detach_dp_input and g is not None else g, - reverse=True, - noise_scale=self.inference_noise_scale_dp, - lang_emb=lang_emb.detach() if self.args.detach_dp_input and lang_emb is not None else lang_emb - ) else: attn_log_durations = torch.log(attn_durations + 1e-6) * x_mask log_durations = self.duration_predictor( @@ -1109,10 +1095,9 @@ class Vits(BaseTTS): g=g.detach() if self.args.detach_dp_input and g is not None else g, lang_emb=lang_emb.detach() if self.args.detach_dp_input and lang_emb is not None else lang_emb, ) - predicted_durations = log_durations loss_duration = torch.sum((log_durations - attn_log_durations) ** 2, [1, 2]) / torch.sum(x_mask) outputs["loss_duration"] = loss_duration - return outputs, attn, predicted_durations + return outputs, attn def upsampling_z(self, z, slice_ids=None, y_lengths=None, y_mask=None): spec_segment_size = self.spec_segment_size @@ -1268,7 +1253,7 @@ class Vits(BaseTTS): else: g_dp = torch.cat([g_dp, pros_emb], dim=1) # [b, h1+h2, 1] - outputs, attn, predicted_durations = self.forward_mas(outputs, z_p, m_p, logs_p, x, x_mask, y_mask, g=g_dp, lang_emb=lang_emb) + outputs, attn = self.forward_mas(outputs, z_p, m_p, logs_p, x, x_mask, y_mask, g=g_dp, lang_emb=lang_emb) # expand prior m_p_expanded = torch.einsum("klmn, kjm -> kjn", [attn, m_p]) @@ -1322,43 +1307,6 @@ class Vits(BaseTTS): else: gt_cons_emb, syn_cons_emb = None, None - end2end_dict = None - if self.args.use_end2end_loss: - # predicted_durations - w = torch.exp(predicted_durations) * x_mask * self.length_scale - w_ceil = torch.ceil(w) - y_lengths_end2end = torch.clamp_min(torch.sum(w_ceil, [1, 2]), 1).long() - y_mask_end2end = sequence_mask(y_lengths_end2end, None).to(x_mask.dtype).unsqueeze(1) # [B, 1, T_dec] - - attn_mask = x_mask * y_mask_end2end.transpose(1, 2) # [B, 1, T_enc] * [B, T_dec, 1] - attn_end2end = generate_path(w_ceil.squeeze(1), attn_mask.squeeze(1).transpose(1, 2)) - - m_p_end2end = torch.matmul(attn_end2end.transpose(1, 2), m_p.transpose(1, 2)).transpose(1, 2) - logs_p_end2end = torch.matmul(attn_end2end.transpose(1, 2), logs_p.transpose(1, 2)).transpose(1, 2) - - z_p_end2end = m_p_end2end * y_mask_end2end #+ torch.randn_like(m_p_end2end) * torch.exp(logs_p_end2end) * self.inference_noise_scale - - # conditional module - if self.args.use_prosody_conditional_flow_module: - if self.args.prosody_conditional_flow_module_on_decoder: - z_p_end2end = self.prosody_conditional_module(z_p_end2end, y_mask_end2end, g=eg if (self.args.use_emotion_embedding or self.args.use_external_emotions_embeddings) else pros_emb, reverse=True) - - z_end2end = self.flow(z_p_end2end, y_mask_end2end, g=g, reverse=True) - - # interpolate z if needed - z_end2end, _, _, y_mask_end2end = self.upsampling_z(z, y_lengths=y_lengths_end2end, y_mask=y_mask_end2end) - # z_slice_end2end, spec_segment_size, slice_ids_end2end, _ = self.upsampling_z(z_slice_end2end, slice_ids=slice_ids_end2end) - - # generate all z using the vocoder - o_end2end = self.waveform_decoder(z_end2end, g=g) - wav_seg_end2end = waveform - - z_predicted_gt_durations = None - if not self.args.use_soft_dtw: - z_predicted_gt_durations = self.flow(m_p_expanded * y_mask, y_mask, g=g, reverse=True) - - end2end_dict = {"logs_p_end2end": logs_p_end2end, "logs_p": logs_p_expanded, "logs_q": logs_q, "z_mask": y_mask, "z_mask_end2end": y_mask_end2end, "z": z, "z_predicted_end2end": z_end2end, "z_predicted": z_predicted_gt_durations, "model_outputs": o_end2end, "waveform_seg": wav_seg_end2end} - outputs.update( { "model_outputs": o, @@ -1377,8 +1325,7 @@ class Vits(BaseTTS): "loss_prosody_enc_spk_rev_classifier": l_pros_speaker, "loss_prosody_enc_emo_classifier": l_pros_emotion, "loss_text_enc_spk_rev_classifier": l_text_speaker, - "loss_text_enc_emo_classifier": l_text_emotion, - "end2end_info": end2end_dict, + "loss_text_enc_emo_classifier": l_text_emotion } ) return outputs @@ -1686,21 +1633,13 @@ class Vits(BaseTTS): outputs["model_outputs"].detach(), outputs["waveform_seg"], outputs["m_p"].detach(), outputs["z_p"].detach() ) - end2end_info = None - if self.args.use_end2end_loss: - scores_disc_fake_end2end, _, scores_disc_real_end2end, _, _, _, _, _ = self.disc( - outputs["end2end_info"]["model_outputs"].detach(), self.model_outputs_cache["end2end_info"]["waveform_seg"] - ) - end2end_info = {"scores_disc_real": scores_disc_real_end2end, "scores_disc_fake": scores_disc_fake_end2end} - # compute loss with autocast(enabled=False): # use float32 for the criterion loss_dict = criterion[optimizer_idx]( scores_disc_real, scores_disc_fake, scores_disc_zp, - scores_disc_mp, - end2end_info=end2end_info, + scores_disc_mp ) return outputs, loss_dict @@ -1735,14 +1674,6 @@ class Vits(BaseTTS): self.model_outputs_cache["model_outputs"], self.model_outputs_cache["waveform_seg"], self.model_outputs_cache["m_p"], self.model_outputs_cache["z_p"].detach() ) - if self.args.use_end2end_loss: - scores_disc_fake_end2end, feats_disc_fake_end2end, _, feats_disc_real_end2end, _, _, _, _, _ = self.disc( - self.model_outputs_cache["end2end_info"]["model_outputs"], self.model_outputs_cache["end2end_info"]["waveform_seg"] - ) - self.model_outputs_cache["end2end_info"]["scores_disc_fake"] = scores_disc_fake_end2end - self.model_outputs_cache["end2end_info"]["feats_disc_fake"] = feats_disc_fake_end2end - self.model_outputs_cache["end2end_info"]["feats_disc_real"] = feats_disc_real_end2end - # compute losses with autocast(enabled=False): # use float32 for the criterion loss_dict = criterion[optimizer_idx]( @@ -1768,8 +1699,7 @@ class Vits(BaseTTS): loss_text_enc_emo_classifier=self.model_outputs_cache["loss_text_enc_emo_classifier"], scores_disc_mp=scores_disc_mp, feats_disc_mp=feats_disc_mp, - feats_disc_zp=feats_disc_zp, - end2end_info=self.model_outputs_cache["end2end_info"], + feats_disc_zp=feats_disc_zp ) return self.model_outputs_cache, loss_dict