From 1b6f8d0e46ca6a39a97dfd63af548e540a5377ca Mon Sep 17 00:00:00 2001 From: Edresson Casanova Date: Mon, 6 Nov 2023 14:52:09 -0300 Subject: [PATCH] Update unit tests and recipes --- TTS/tts/models/xtts.py | 2 +- recipes/ljspeech/xtts_v1/train_gpt_xtts.py | 4 ++-- recipes/ljspeech/xtts_v2/train_gpt_xtts.py | 4 ++-- tests/xtts_tests/test_xtts_gpt_train.py | 2 +- tests/xtts_tests/test_xtts_v2-0_gpt_train.py | 2 +- 5 files changed, 7 insertions(+), 7 deletions(-) diff --git a/TTS/tts/models/xtts.py b/TTS/tts/models/xtts.py index fdaeb7de..e70799ec 100644 --- a/TTS/tts/models/xtts.py +++ b/TTS/tts/models/xtts.py @@ -433,7 +433,7 @@ class Xtts(BaseTTS): ): # deal with multiples references if not isinstance(audio_path, list): - audio_paths = list(audio_path) + audio_paths = [audio_path] else: audio_paths = audio_path diff --git a/recipes/ljspeech/xtts_v1/train_gpt_xtts.py b/recipes/ljspeech/xtts_v1/train_gpt_xtts.py index 268a0335..a7ae40e2 100644 --- a/recipes/ljspeech/xtts_v1/train_gpt_xtts.py +++ b/recipes/ljspeech/xtts_v1/train_gpt_xtts.py @@ -71,9 +71,9 @@ if not os.path.isfile(TOKENIZER_FILE) or not os.path.isfile(XTTS_CHECKPOINT): # Training sentences generations -SPEAKER_REFERENCE = ( +SPEAKER_REFERENCE = [ "./tests/data/ljspeech/wavs/LJ001-0002.wav" # speaker reference to be used in training test sentences -) +] LANGUAGE = config_dataset.language diff --git a/recipes/ljspeech/xtts_v2/train_gpt_xtts.py b/recipes/ljspeech/xtts_v2/train_gpt_xtts.py index 47a52d57..989b1936 100644 --- a/recipes/ljspeech/xtts_v2/train_gpt_xtts.py +++ b/recipes/ljspeech/xtts_v2/train_gpt_xtts.py @@ -71,9 +71,9 @@ if not os.path.isfile(TOKENIZER_FILE) or not os.path.isfile(XTTS_CHECKPOINT): # Training sentences generations -SPEAKER_REFERENCE = ( +SPEAKER_REFERENCE = [ "./tests/data/ljspeech/wavs/LJ001-0002.wav" # speaker reference to be used in training test sentences -) +] LANGUAGE = config_dataset.language diff --git a/tests/xtts_tests/test_xtts_gpt_train.py b/tests/xtts_tests/test_xtts_gpt_train.py index 03514daa..12c547d6 100644 --- a/tests/xtts_tests/test_xtts_gpt_train.py +++ b/tests/xtts_tests/test_xtts_gpt_train.py @@ -60,7 +60,7 @@ XTTS_CHECKPOINT = None # "/raid/edresson/dev/Checkpoints/XTTS_evaluation/xtts_s # Training sentences generations -SPEAKER_REFERENCE = "tests/data/ljspeech/wavs/LJ001-0002.wav" # speaker reference to be used in training test sentences +SPEAKER_REFERENCE = ["tests/data/ljspeech/wavs/LJ001-0002.wav"] # speaker reference to be used in training test sentences LANGUAGE = config_dataset.language diff --git a/tests/xtts_tests/test_xtts_v2-0_gpt_train.py b/tests/xtts_tests/test_xtts_v2-0_gpt_train.py index 80995038..3d6ef60e 100644 --- a/tests/xtts_tests/test_xtts_v2-0_gpt_train.py +++ b/tests/xtts_tests/test_xtts_v2-0_gpt_train.py @@ -58,7 +58,7 @@ XTTS_CHECKPOINT = None # "/raid/edresson/dev/Checkpoints/XTTS_evaluation/xtts_s # Training sentences generations -SPEAKER_REFERENCE = "tests/data/ljspeech/wavs/LJ001-0002.wav" # speaker reference to be used in training test sentences +SPEAKER_REFERENCE = ["tests/data/ljspeech/wavs/LJ001-0002.wav"] # speaker reference to be used in training test sentences LANGUAGE = config_dataset.language