MCPcopy
hub / github.com/coqui-ai/TTS / test_xtts

Function test_xtts

tests/zoo_tests/test_models.py:77–95  ·  view source on GitHub ↗

XTTS is too big to run on github actions. We need to test it locally

()

Source from the content-addressed store, hash-verified

75
76
77def test_xtts():
78 """XTTS is too big to run on github actions. We need to test it locally"""
79 output_path = os.path.join(get_tests_output_path(), "output.wav")
80 speaker_wav = os.path.join(get_tests_data_path(), "ljspeech", "wavs", "LJ001-0001.wav")
81 use_gpu = torch.cuda.is_available()
82 if use_gpu:
83 run_cli(
84 "yes | "
85 f"tts --model_name tts_models/multilingual/multi-dataset/xtts_v1.1 "
86 f'--text "This is an example." --out_path "{output_path}" --progress_bar False --use_cuda True '
87 f'--speaker_wav "{speaker_wav}" --language_idx "en"'
88 )
89 else:
90 run_cli(
91 "yes | "
92 f"tts --model_name tts_models/multilingual/multi-dataset/xtts_v1.1 "
93 f'--text "This is an example." --out_path "{output_path}" --progress_bar False '
94 f'--speaker_wav "{speaker_wav}" --language_idx "en"'
95 )
96
97
98def test_xtts_streaming():

Callers

nothing calls this directly

Calls 4

get_tests_output_pathFunction · 0.90
get_tests_data_pathFunction · 0.90
run_cliFunction · 0.90
is_availableMethod · 0.45

Tested by

no test coverage detected

Used in the wild real call sites across dependent graphs

searching dependent graphs…