From 2a2a0864c15ba7e7369d07a300a0cc56302ee42e Mon Sep 17 00:00:00 2001 From: Alexander Korolev Date: Fri, 14 Aug 2020 09:38:53 +0200 Subject: [PATCH 1/3] Update train_tts.py align style_input with the new config key "gst_style_input" --- mozilla_voice_tts/bin/train_tts.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mozilla_voice_tts/bin/train_tts.py b/mozilla_voice_tts/bin/train_tts.py index 2b6cbfd0..80fbff95 100644 --- a/mozilla_voice_tts/bin/train_tts.py +++ b/mozilla_voice_tts/bin/train_tts.py @@ -441,7 +441,7 @@ def evaluate(model, criterion, ap, global_step, epoch, speaker_mapping=None): test_figures = {} print(" | > Synthesizing test sentences") speaker_id = 0 if c.use_speaker_embedding else None - style_wav = c.get("style_wav_for_test") + style_wav = c.get("gst_style_input") for idx, test_sentence in enumerate(test_sentences): try: wav, alignment, decoder_output, postnet_output, stop_tokens, _ = synthesis( From c90fce8d8c98fc58d9ab329a2ca95bc2539d4061 Mon Sep 17 00:00:00 2001 From: thllwg Date: Fri, 14 Aug 2020 11:06:32 +0200 Subject: [PATCH 2/3] Fix for #497 --- mozilla_voice_tts/bin/train_tts.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mozilla_voice_tts/bin/train_tts.py b/mozilla_voice_tts/bin/train_tts.py index 2b6cbfd0..9fef0888 100644 --- a/mozilla_voice_tts/bin/train_tts.py +++ b/mozilla_voice_tts/bin/train_tts.py @@ -100,7 +100,7 @@ def format_data(data, speaker_mapping=None): if c.use_speaker_embedding: if c.use_external_speaker_embedding_file: - speaker_embeddings = data[8] + speaker_embeddings = data[8] speaker_ids = None else: speaker_ids = [ @@ -624,7 +624,7 @@ def main(args): # pylint: disable=redefined-outer-name train_avg_loss_dict, global_step = train(model, criterion, optimizer, optimizer_st, scheduler, ap, global_step, epoch, amp, speaker_mapping) - eval_avg_loss_dict = evaluate(model, criterion, ap, global_step, epoch) + eval_avg_loss_dict = evaluate(model, criterion, ap, global_step, epoch, speaker_mapping) c_logger.print_epoch_end(epoch, eval_avg_loss_dict) target_loss = train_avg_loss_dict['avg_postnet_loss'] if c.run_eval: From e07ff8abe8f66e60b07f864f85d08e488d8a988a Mon Sep 17 00:00:00 2001 From: thllwg Date: Fri, 14 Aug 2020 11:12:37 +0200 Subject: [PATCH 3/3] remove empty space --- mozilla_voice_tts/bin/train_tts.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mozilla_voice_tts/bin/train_tts.py b/mozilla_voice_tts/bin/train_tts.py index 9fef0888..f8700a3e 100644 --- a/mozilla_voice_tts/bin/train_tts.py +++ b/mozilla_voice_tts/bin/train_tts.py @@ -100,7 +100,7 @@ def format_data(data, speaker_mapping=None): if c.use_speaker_embedding: if c.use_external_speaker_embedding_file: - speaker_embeddings = data[8] + speaker_embeddings = data[8] speaker_ids = None else: speaker_ids = [