From ecf327a118bc907d2f1a36743a88d4f157af4109 Mon Sep 17 00:00:00 2001
From: Edresson <edresson1@gmail.com>
Date: Sat, 28 Aug 2021 09:59:48 -0300
Subject: [PATCH] Add VITS multispeaker train unit test

---
 .../tts_tests/test_vits_speaker_emb_train.py  | 63 +++++++++++++++++++
 1 file changed, 63 insertions(+)
 create mode 100644 tests/tts_tests/test_vits_speaker_emb_train.py

diff --git a/tests/tts_tests/test_vits_speaker_emb_train.py b/tests/tts_tests/test_vits_speaker_emb_train.py
new file mode 100644
index 00000000..7028a983
--- /dev/null
+++ b/tests/tts_tests/test_vits_speaker_emb_train.py
@@ -0,0 +1,63 @@
+import glob
+import os
+import shutil
+
+from tests import get_device_id, get_tests_output_path, run_cli
+from TTS.tts.configs import VitsConfig
+
+config_path = os.path.join(get_tests_output_path(), "test_model_config.json")
+output_path = os.path.join(get_tests_output_path(), "train_outputs")
+
+
+config = VitsConfig(
+    batch_size=2,
+    eval_batch_size=2,
+    num_loader_workers=0,
+    num_eval_loader_workers=0,
+    text_cleaner="english_cleaners",
+    use_phonemes=True,
+    use_espeak_phonemes=True,
+    phoneme_language="en-us",
+    phoneme_cache_path="tests/data/ljspeech/phoneme_cache/",
+    run_eval=True,
+    test_delay_epochs=-1,
+    epochs=1,
+    print_step=1,
+    print_eval=True,
+    test_sentences=[
+        ["Be a voice, not an echo.", "ljspeech"],
+    ],
+)
+# set audio config
+config.audio.do_trim_silence = True
+config.audio.trim_db = 60
+
+# active multispeaker d-vec mode
+config.model_args.use_speaker_embedding = True
+config.model_args.use_d_vector_file = False
+config.model_args.d_vector_file = None
+config.model_args.d_vector_dim = 256
+
+
+config.save_json(config_path)
+
+# train the model for one epoch
+command_train = (
+    f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_tts.py --config_path {config_path} "
+    f"--coqpit.output_path {output_path} "
+    "--coqpit.datasets.0.name ljspeech "
+    "--coqpit.datasets.0.meta_file_train metadata.csv "
+    "--coqpit.datasets.0.meta_file_val metadata.csv "
+    "--coqpit.datasets.0.path tests/data/ljspeech "
+    "--coqpit.datasets.0.meta_file_attn_mask tests/data/ljspeech/metadata_attn_mask.txt "
+    "--coqpit.test_delay_epochs 0"
+)
+run_cli(command_train)
+
+# Find latest folder
+continue_path = max(glob.glob(os.path.join(output_path, "*/")), key=os.path.getmtime)
+
+# restore the model and continue training for one more epoch
+command_train = f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_tts.py --continue_path {continue_path} "
+run_cli(command_train)
+shutil.rmtree(continue_path)