Skip to content

VoiceStudio is the open-source, fully-local ElevenLabs alternative — voice cloning, voice design, video dubbing, dictation, transcription & audiobook creation in 646 languages.

2.6 KB
Raw Blame
"""Desktop language names reach the Supertonic SDK as ISO language codes."""
import importlib.util
import io
from pathlib import Path

import pytest

LANGUAGES = [
    ('English', 'en'), ('Korean', 'ko'), ('Japanese', 'ja'), ('Arabic', 'ar'),
    ('Bulgarian', 'bg'), ('Czech', 'cs'), ('Danish', 'da'), ('German', 'de'),
    ('Greek', 'el'), ('Spanish', 'es'), ('Estonian', 'et'), ('Finnish', 'fi'),
    ('French', 'fr'), ('Hindi', 'hi'), ('Croatian', 'hr'), ('Hungarian', 'hu'),
    ('Indonesian', 'id'), ('Italian', 'it'), ('Lithuanian', 'lt'), ('Latvian', 'lv'),
    ('Dutch', 'nl'), ('Polish', 'pl'), ('Portuguese', 'pt'), ('Romanian', 'ro'),
    ('Russian', 'ru'), ('Slovak', 'sk'), ('Slovenian', 'sl'), ('Swedish', 'sv'),
    ('Turkish', 'tr'), ('Ukrainian', 'uk'), ('Vietnamese', 'vi'),
]


@pytest.fixture
def sidecar():
    path = Path(__file__).resolve().parents[1] / 'backend/engines/supertonic3/sidecar.py'
    spec = importlib.util.spec_from_file_location('supertonic_language_test', path)
    module = importlib.util.module_from_spec(spec)
    spec.loader.exec_module(module)
    return module


@pytest.mark.parametrize('name,code', LANGUAGES)
def test_synthesis_dispatches_full_language_names_to_sdk(sidecar, monkeypatch, name, code):
    calls = []

    class SDK:
        sample_rate = 44100

        def get_voice_style(self, voice_name):
            assert voice_name == 'M1'
            return 'test-style'

        def synthesize(self, **kwargs):
            calls.append(kwargs)
            assert kwargs['lang'] == code
            return object(), 0.1

    monkeypatch.setattr(sidecar, '_load_tts', lambda stdout: SDK())
    monkeypatch.setattr(sidecar, '_wav_float_to_pcm_b64', lambda wav, sr: ('AAA=', sr, 1))
    frames = []
    monkeypatch.setattr(sidecar, '_send', lambda stdout, frame: frames.append(frame))
    sidecar._handle_synthesize({'text': 'A language test.', 'lang': name}, io.BytesIO())
    assert len(calls) == 1
    assert frames[0]['op'] == 'audio'


@pytest.mark.parametrize('name,code', LANGUAGES)
def test_language_codes_remain_accepted(sidecar, name, code):
    assert sidecar._normalize_lang(code.upper()) == code


@pytest.mark.parametrize('raw,expected', [
    (None, 'na'), ('', 'na'), ('  AUTO ', 'na'), (42, 'na'), ('na', 'na'),
    ('pt-BR', 'pt'), ('PT_pt', 'pt'), ('en-US', 'en'),
    ('  Portuguese  ', 'pt'), ('eStOnIaN', 'et'),
    ('Englishish', 'englishish'), ('Polishish', 'polishish'),
    ('English-unknown', 'english-unknown'), ('en-unknown', 'en-unknown'),
    ('zz-US', 'zz-us'), ('es-419', 'es'),
])
def test_language_normalization_controls(sidecar, raw, expected):
    assert sidecar._normalize_lang(raw) == expected