From 4b16080f4b3d45fc7132fc69a6b1e5f23d229719 Mon Sep 17 00:00:00 2001 From: PR Bot Date: Tue, 17 Mar 2026 18:17:54 +0800 Subject: [PATCH] feat: add MiniMax TTS as a new dubbing provider Add MiniMax Cloud TTS (speech-2.8-hd / speech-2.8-turbo) as a new TTS provider for video dubbing, with 12 built-in voices, Streamlit sidebar integration, full i18n support (7 locales), and unit + integration tests. - New provider: core/tts_backend/minimax_tts.py - Config: minimax_tts section in config.yaml - UI: voice/model selectors in sidebar settings - Docs: TTS comparison tables updated (EN + ZH) - Tests: 8 unit tests + 3 integration tests Co-Authored-By: Octopus --- config.yaml | 8 +- core/st_utils/sidebar_setting.py | 37 ++++- core/tts_backend/minimax_tts.py | 79 +++++++++ core/tts_backend/tts_main.py | 3 + docs/pages/docs/start.en-US.md | 2 + docs/pages/docs/start.zh-CN.md | 2 + tests/__init__.py | 0 tests/test_minimax_tts.py | 272 +++++++++++++++++++++++++++++++ translations/en.json | 17 +- translations/es.json | 17 +- translations/fr.json | 17 +- translations/ja.json | 17 +- translations/ru.json | 17 +- translations/zh-CN.json | 17 +- translations/zh-HK.json | 17 +- 15 files changed, 513 insertions(+), 9 deletions(-) create mode 100644 core/tts_backend/minimax_tts.py create mode 100644 tests/__init__.py create mode 100644 tests/test_minimax_tts.py diff --git a/config.yaml b/config.yaml index 099b4bdb..bf4ea2a6 100644 --- a/config.yaml +++ b/config.yaml @@ -70,7 +70,7 @@ reflect_translate: true pause_before_translate: false ## ======================== Dubbing Settings ======================== ## -# TTS selection [sf_fish_tts, openai_tts, gpt_sovits, azure_tts, fish_tts, edge_tts, custom_tts] +# TTS selection [sf_fish_tts, openai_tts, gpt_sovits, azure_tts, fish_tts, edge_tts, custom_tts, minimax_tts] tts_method: 'azure_tts' # SiliconFlow FishTTS @@ -119,6 +119,12 @@ gpt_sovits: f5tts: 302_api: 'YOUR_302_API_KEY' +# MiniMax TTS configuration, get API key from https://www.minimax.io +minimax_tts: + api_key: 'YOUR_MINIMAX_API_KEY' + model: 'speech-2.8-hd' + voice: 'English_Graceful_Lady' + # *Audio speed range speed_factor: min: 1 diff --git a/core/st_utils/sidebar_setting.py b/core/st_utils/sidebar_setting.py index 5c5d24aa..42f13dae 100644 --- a/core/st_utils/sidebar_setting.py +++ b/core/st_utils/sidebar_setting.py @@ -84,7 +84,7 @@ def page_setting(): update_key("burn_subtitles", burn_subtitles) st.rerun() with st.expander(t("Dubbing Settings"), expanded=True): - tts_methods = ["azure_tts", "openai_tts", "fish_tts", "sf_fish_tts", "edge_tts", "gpt_sovits", "custom_tts", "sf_cosyvoice2", "f5tts"] + tts_methods = ["azure_tts", "openai_tts", "fish_tts", "sf_fish_tts", "edge_tts", "gpt_sovits", "custom_tts", "sf_cosyvoice2", "f5tts", "minimax_tts"] select_tts = st.selectbox(t("TTS Method"), options=tts_methods, index=tts_methods.index(load_key("tts_method"))) if select_tts != load_key("tts_method"): update_key("tts_method", select_tts) @@ -151,6 +151,41 @@ def page_setting(): elif select_tts == "f5tts": config_input("302ai API", "f5tts.302_api") + + elif select_tts == "minimax_tts": + config_input(t("MiniMax API Key"), "minimax_tts.api_key") + minimax_voices = { + "English_Graceful_Lady": t("English Female, Graceful"), + "English_Insightful_Speaker": t("English Male, Insightful"), + "English_radiant_girl": t("English Female, Radiant"), + "English_Persuasive_Man": t("English Male, Persuasive"), + "English_Lucky_Robot": t("English, Robot"), + "Wise_Woman": t("Female, Wise"), + "Friendly_Person": t("Friendly"), + "Inspirational_girl": t("Female, Inspirational"), + "Deep_Voice_Man": t("Male, Deep Voice"), + "sweet_girl": t("Female, Sweet"), + "cute_boy": t("Male, Cute"), + "lovely_girl": t("Female, Lovely"), + } + selected_voice = st.selectbox( + t("MiniMax Voice"), + options=list(minimax_voices.keys()), + format_func=lambda x: minimax_voices[x], + index=list(minimax_voices.keys()).index(load_key("minimax_tts.voice")) if load_key("minimax_tts.voice") in minimax_voices else 0 + ) + if selected_voice != load_key("minimax_tts.voice"): + update_key("minimax_tts.voice", selected_voice) + st.rerun() + minimax_models = ["speech-2.8-hd", "speech-2.8-turbo"] + selected_model = st.selectbox( + t("MiniMax TTS Model"), + options=minimax_models, + index=minimax_models.index(load_key("minimax_tts.model")) if load_key("minimax_tts.model") in minimax_models else 0 + ) + if selected_model != load_key("minimax_tts.model"): + update_key("minimax_tts.model", selected_model) + st.rerun() def check_api(): try: diff --git a/core/tts_backend/minimax_tts.py b/core/tts_backend/minimax_tts.py new file mode 100644 index 00000000..892a0a5f --- /dev/null +++ b/core/tts_backend/minimax_tts.py @@ -0,0 +1,79 @@ +import io +import requests +from pathlib import Path +from pydub import AudioSegment +from core.utils import load_key, except_handler + +BASE_URL = "https://api.minimax.io/v1/t2a_v2" +BASE_URL_CN = "https://api.minimaxi.com/v1/t2a_v2" + +VOICE_LIST = [ + "English_Graceful_Lady", + "English_Insightful_Speaker", + "English_radiant_girl", + "English_Persuasive_Man", + "English_Lucky_Robot", + "Wise_Woman", + "Friendly_Person", + "Inspirational_girl", + "Deep_Voice_Man", + "sweet_girl", + "cute_boy", + "lovely_girl", +] + +MODEL_LIST = ["speech-2.8-hd", "speech-2.8-turbo"] + +# refer to: https://platform.minimax.io/docs/api-reference/speech-t2a-http + +@except_handler("Failed to generate audio using MiniMax TTS", retry=3, delay=1) +def minimax_tts(text, save_path): + API_KEY = load_key("minimax_tts.api_key") + voice = load_key("minimax_tts.voice") + model = load_key("minimax_tts.model") + + if voice not in VOICE_LIST: + raise ValueError(f"Invalid voice: {voice}. Please choose from {VOICE_LIST}") + + payload = { + "model": model, + "text": text, + "stream": False, + "voice_setting": { + "voice_id": voice, + "speed": 1.0, + "vol": 1.0, + "pitch": 0, + }, + "audio_setting": { + "format": "mp3", + "sample_rate": 32000, + }, + } + + headers = { + "Authorization": f"Bearer {API_KEY}", + "Content-Type": "application/json", + } + + response = requests.post(BASE_URL, headers=headers, json=payload, timeout=60) + response.raise_for_status() + + result = response.json() + if "data" not in result or "audio" not in result["data"]: + raise ValueError(f"Unexpected API response: {result}") + + audio_hex = result["data"]["audio"] + audio_bytes = bytes.fromhex(audio_hex) + + # Convert mp3 to wav using pydub + speech_file_path = Path(save_path) + speech_file_path.parent.mkdir(parents=True, exist_ok=True) + + audio = AudioSegment.from_mp3(io.BytesIO(audio_bytes)) + audio.export(save_path, format="wav") + print(f"Audio saved to {speech_file_path}") + + +if __name__ == "__main__": + minimax_tts("Hi! Welcome to VideoLingo!", "test.wav") diff --git a/core/tts_backend/tts_main.py b/core/tts_backend/tts_main.py index 49cc0c29..bcf56515 100644 --- a/core/tts_backend/tts_main.py +++ b/core/tts_backend/tts_main.py @@ -11,6 +11,7 @@ from core.tts_backend.edge_tts import edge_tts from core.tts_backend.sf_cosyvoice2 import cosyvoice_tts_for_videolingo from core.tts_backend.custom_tts import custom_tts +from core.tts_backend.minimax_tts import minimax_tts from core.prompts import get_correct_text_prompt from core.tts_backend._302_f5tts import f5_tts_for_videolingo from core.utils import * @@ -64,6 +65,8 @@ def tts_main(text, save_as, number, task_df): cosyvoice_tts_for_videolingo(text, save_as, number, task_df) elif TTS_METHOD == 'f5tts': f5_tts_for_videolingo(text, save_as, number, task_df) + elif TTS_METHOD == 'minimax_tts': + minimax_tts(text, save_as) # Check generated audio duration duration = get_audio_duration(save_as) diff --git a/docs/pages/docs/start.en-US.md b/docs/pages/docs/start.en-US.md index 8de34745..7b0a1835 100644 --- a/docs/pages/docs/start.en-US.md +++ b/docs/pages/docs/start.en-US.md @@ -34,9 +34,11 @@ VideoLingo provides multiple TTS integration methods. Here's a comparison (skip | 🎙️ SiliconFlow FishTTS | [SiliconFlow](https://cloud.siliconflow.cn/i/ttKDEsxE) | Voice Clone | Unstable cloning effect | 😃 | 😃 | | 🗣 Edge TTS | Local | Completely free | Average effect | 😐 | 😐 | | 🗣️ GPT-SoVITS | Local | Best voice cloning | Only supports Chinese/English, requires local inference, complex setup | 🏆 | 🚫 | +| 🔊 MiniMax TTS | [MiniMax](https://www.minimax.io) | HD quality, many voices | Requires API key | 😃 | 🤩 | - For SiliconFlow FishTTS, get key from [SiliconFlow](https://cloud.siliconflow.cn/i/ttKDEsxE), note that cloning feature requires paid credits; - For OpenAI TTS, Azure TTS, and Fish TTS, use [302AI](https://gpt302.saaslink.net/C2oHR9) - one API key provides access to all three services +- For MiniMax TTS, get API key from [MiniMax](https://www.minimax.io), supports `speech-2.8-hd` (recommended) and `speech-2.8-turbo` models > Wanna use your own TTS? Modify in `core/all_tts_functions/custom_tts.py`!
diff --git a/docs/pages/docs/start.zh-CN.md b/docs/pages/docs/start.zh-CN.md index 972c84e7..cf29b7a5 100644 --- a/docs/pages/docs/start.zh-CN.md +++ b/docs/pages/docs/start.zh-CN.md @@ -34,9 +34,11 @@ VideoLingo提供了多种 tts 接入方式，以下是对比（如不使用配 | 🎙️ SiliconFlow FishTTS | [硅基流动](https://cloud.siliconflow.cn/i/ttKDEsxE) | 语音克隆 | 克隆效果不稳定 | 😃 | 😃 | | 🗣 Edge TTS | 本地 | 完全免费 | 效果一般 | 😐 | 😐 | | 🗣️ GPT-SoVITS | 本地 | 最强语音克隆 | 只支持中英文，需要本地训练推理，配置麻烦 | 🏆 | 🚫 | +| 🔊 MiniMax TTS | [MiniMax](https://www.minimax.io) | 高清音质，多种音色 | 需要API密钥 | 😃 | 🤩 | - SiliconFlow FishTTS 请在 [硅基流动](https://cloud.siliconflow.cn/i/ttKDEsxE) 获取key，注意克隆功能需要付费充值积分； - OpenAI TTS、Azure TTS 和 Fish TTS，仅支持 [302AI](https://gpt302.saaslink.net/C2oHR9) - 一个 API key 即可使用所有服务 +- MiniMax TTS 请在 [MiniMax](https://www.minimax.io) 获取API密钥，支持 `speech-2.8-hd`（推荐）和 `speech-2.8-turbo` 模型 > 现在还可以在 `core/all_tts_functions/custom_tts.py` 里自定义tts渠道！
diff --git a/tests/__init__.py b/tests/__init__.py new file mode 100644 index 00000000..e69de29b diff --git a/tests/test_minimax_tts.py b/tests/test_minimax_tts.py new file mode 100644 index 00000000..b7895643 --- /dev/null +++ b/tests/test_minimax_tts.py @@ -0,0 +1,272 @@ +"""Unit and integration tests for MiniMax TTS provider.""" + +import io +import json +import os +import tempfile +import unittest +from unittest.mock import patch, MagicMock + +from pydub import AudioSegment + + +class TestMiniMaxTTSUnit(unittest.TestCase): + """Unit tests for MiniMax TTS provider (no real API calls).""" + + def _make_fake_mp3_hex(self): + """Create a minimal valid MP3 audio and return its hex string.""" + silence = AudioSegment.silent(duration=100) # 100ms + buf = io.BytesIO() + silence.export(buf, format="mp3") + return buf.getvalue().hex() + + @patch("core.tts_backend.minimax_tts.requests.post") + @patch("core.tts_backend.minimax_tts.load_key") + def test_successful_tts_generation(self, mock_load_key, mock_post): + """Test that minimax_tts generates a WAV file on successful API response.""" + from core.tts_backend.minimax_tts import minimax_tts + + mock_load_key.side_effect = lambda key: { + "minimax_tts.api_key": "test-key", + "minimax_tts.voice": "English_Graceful_Lady", + "minimax_tts.model": "speech-2.8-hd", + }[key] + + fake_hex = self._make_fake_mp3_hex() + mock_response = MagicMock() + mock_response.status_code = 200 + mock_response.json.return_value = {"data": {"audio": fake_hex}} + mock_response.raise_for_status = MagicMock() + mock_post.return_value = mock_response + + with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f: + save_path = f.name + + try: + minimax_tts("Hello world", save_path) + self.assertTrue(os.path.exists(save_path)) + self.assertGreater(os.path.getsize(save_path), 0) + + # Verify it's a valid audio file + audio = AudioSegment.from_wav(save_path) + self.assertGreater(len(audio), 0) + finally: + if os.path.exists(save_path): + os.remove(save_path) + + @patch("core.tts_backend.minimax_tts.requests.post") + @patch("core.tts_backend.minimax_tts.load_key") + def test_correct_api_request_params(self, mock_load_key, mock_post): + """Test that the API is called with correct parameters.""" + from core.tts_backend.minimax_tts import minimax_tts + + mock_load_key.side_effect = lambda key: { + "minimax_tts.api_key": "sk-test-123", + "minimax_tts.voice": "English_Insightful_Speaker", + "minimax_tts.model": "speech-2.8-turbo", + }[key] + + fake_hex = self._make_fake_mp3_hex() + mock_response = MagicMock() + mock_response.status_code = 200 + mock_response.json.return_value = {"data": {"audio": fake_hex}} + mock_response.raise_for_status = MagicMock() + mock_post.return_value = mock_response + + with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f: + save_path = f.name + + try: + minimax_tts("Test text", save_path) + + mock_post.assert_called_once() + call_kwargs = mock_post.call_args + self.assertEqual(call_kwargs.kwargs["headers"]["Authorization"], "Bearer sk-test-123") + payload = call_kwargs.kwargs["json"] + self.assertEqual(payload["model"], "speech-2.8-turbo") + self.assertEqual(payload["text"], "Test text") + self.assertEqual(payload["voice_setting"]["voice_id"], "English_Insightful_Speaker") + self.assertFalse(payload["stream"]) + finally: + if os.path.exists(save_path): + os.remove(save_path) + + @patch("core.tts_backend.minimax_tts.load_key") + def test_invalid_voice_raises_error(self, mock_load_key): + """Test that an invalid voice ID raises ValueError.""" + from core.tts_backend.minimax_tts import minimax_tts + + mock_load_key.side_effect = lambda key: { + "minimax_tts.api_key": "test-key", + "minimax_tts.voice": "NonExistent_Voice", + "minimax_tts.model": "speech-2.8-hd", + }[key] + + with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f: + save_path = f.name + + try: + with self.assertRaises(Exception): + minimax_tts("Hello", save_path) + finally: + if os.path.exists(save_path): + os.remove(save_path) + + @patch("core.tts_backend.minimax_tts.requests.post") + @patch("core.tts_backend.minimax_tts.load_key") + def test_missing_audio_data_raises_error(self, mock_load_key, mock_post): + """Test that missing audio data in response raises ValueError.""" + from core.tts_backend.minimax_tts import minimax_tts + + mock_load_key.side_effect = lambda key: { + "minimax_tts.api_key": "test-key", + "minimax_tts.voice": "English_Graceful_Lady", + "minimax_tts.model": "speech-2.8-hd", + }[key] + + mock_response = MagicMock() + mock_response.status_code = 200 + mock_response.json.return_value = {"error": "some error"} + mock_response.raise_for_status = MagicMock() + mock_post.return_value = mock_response + + with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f: + save_path = f.name + + try: + with self.assertRaises(Exception): + minimax_tts("Hello", save_path) + finally: + if os.path.exists(save_path): + os.remove(save_path) + + @patch("core.tts_backend.minimax_tts.requests.post") + @patch("core.tts_backend.minimax_tts.load_key") + def test_creates_parent_directories(self, mock_load_key, mock_post): + """Test that parent directories are created if they don't exist.""" + from core.tts_backend.minimax_tts import minimax_tts + + mock_load_key.side_effect = lambda key: { + "minimax_tts.api_key": "test-key", + "minimax_tts.voice": "English_Graceful_Lady", + "minimax_tts.model": "speech-2.8-hd", + }[key] + + fake_hex = self._make_fake_mp3_hex() + mock_response = MagicMock() + mock_response.status_code = 200 + mock_response.json.return_value = {"data": {"audio": fake_hex}} + mock_response.raise_for_status = MagicMock() + mock_post.return_value = mock_response + + with tempfile.TemporaryDirectory() as tmpdir: + save_path = os.path.join(tmpdir, "nested", "dir", "output.wav") + minimax_tts("Hello", save_path) + self.assertTrue(os.path.exists(save_path)) + + def test_voice_list_not_empty(self): + """Test that VOICE_LIST contains expected voices.""" + from core.tts_backend.minimax_tts import VOICE_LIST + self.assertGreater(len(VOICE_LIST), 0) + self.assertIn("English_Graceful_Lady", VOICE_LIST) + self.assertIn("Wise_Woman", VOICE_LIST) + + def test_model_list_not_empty(self): + """Test that MODEL_LIST contains expected models.""" + from core.tts_backend.minimax_tts import MODEL_LIST + self.assertIn("speech-2.8-hd", MODEL_LIST) + self.assertIn("speech-2.8-turbo", MODEL_LIST) + + def test_base_url_uses_minimax_io(self): + """Test that default base URL points to api.minimax.io.""" + from core.tts_backend.minimax_tts import BASE_URL + self.assertTrue(BASE_URL.startswith("https://api.minimax.io")) + + +class TestMiniMaxTTSIntegration(unittest.TestCase): + """Integration tests that call the real MiniMax TTS API. + + Requires MINIMAX_API_KEY environment variable. + """ + + @classmethod + def setUpClass(cls): + cls.api_key = os.environ.get("MINIMAX_API_KEY") + if not cls.api_key: + raise unittest.SkipTest("MINIMAX_API_KEY not set, skipping integration tests") + + @patch("core.tts_backend.minimax_tts.load_key") + def test_real_tts_generation(self, mock_load_key): + """Test TTS generation with a real API call.""" + from core.tts_backend.minimax_tts import minimax_tts + + mock_load_key.side_effect = lambda key: { + "minimax_tts.api_key": self.api_key, + "minimax_tts.voice": "English_Graceful_Lady", + "minimax_tts.model": "speech-2.8-hd", + }[key] + + with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f: + save_path = f.name + + try: + minimax_tts("Hello! Welcome to VideoLingo.", save_path) + self.assertTrue(os.path.exists(save_path)) + self.assertGreater(os.path.getsize(save_path), 1000) + + # Verify it's valid audio with reasonable duration + audio = AudioSegment.from_wav(save_path) + duration_ms = len(audio) + self.assertGreater(duration_ms, 500) # At least 0.5 seconds + self.assertLess(duration_ms, 30000) # Less than 30 seconds + finally: + if os.path.exists(save_path): + os.remove(save_path) + + @patch("core.tts_backend.minimax_tts.load_key") + def test_real_tts_chinese_text(self, mock_load_key): + """Test TTS generation with Chinese text using Wise_Woman voice.""" + from core.tts_backend.minimax_tts import minimax_tts + + mock_load_key.side_effect = lambda key: { + "minimax_tts.api_key": self.api_key, + "minimax_tts.voice": "Wise_Woman", + "minimax_tts.model": "speech-2.8-hd", + }[key] + + with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f: + save_path = f.name + + try: + minimax_tts("你好！欢迎使用VideoLingo。", save_path) + self.assertTrue(os.path.exists(save_path)) + self.assertGreater(os.path.getsize(save_path), 1000) + finally: + if os.path.exists(save_path): + os.remove(save_path) + + @patch("core.tts_backend.minimax_tts.load_key") + def test_real_tts_turbo_model(self, mock_load_key): + """Test TTS generation with speech-2.8-turbo model.""" + from core.tts_backend.minimax_tts import minimax_tts + + mock_load_key.side_effect = lambda key: { + "minimax_tts.api_key": self.api_key, + "minimax_tts.voice": "English_Insightful_Speaker", + "minimax_tts.model": "speech-2.8-turbo", + }[key] + + with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f: + save_path = f.name + + try: + minimax_tts("This is a test with the turbo model.", save_path) + self.assertTrue(os.path.exists(save_path)) + self.assertGreater(os.path.getsize(save_path), 1000) + finally: + if os.path.exists(save_path): + os.remove(save_path) + + +if __name__ == "__main__": + unittest.main() diff --git a/translations/en.json b/translations/en.json index fb216bce..f808a52c 100644 --- a/translations/en.json +++ b/translations/en.json @@ -106,5 +106,20 @@ "No NVIDIA GPU detected": "No NVIDIA GPU detected", "No NVIDIA GPU detected or NVIDIA drivers not properly installed": "No NVIDIA GPU detected or NVIDIA drivers not properly installed", "LLM JSON Format Support": "LLM JSON Format Support", - "Enable if your LLM supports JSON mode output": "Enable if your LLM supports JSON mode output" + "Enable if your LLM supports JSON mode output": "Enable if your LLM supports JSON mode output", + "MiniMax API Key": "MiniMax API Key", + "MiniMax Voice": "MiniMax Voice", + "MiniMax TTS Model": "MiniMax TTS Model", + "English Female, Graceful": "English Female, Graceful", + "English Male, Insightful": "English Male, Insightful", + "English Female, Radiant": "English Female, Radiant", + "English Male, Persuasive": "English Male, Persuasive", + "English, Robot": "English, Robot", + "Female, Wise": "Female, Wise", + "Friendly": "Friendly", + "Female, Inspirational": "Female, Inspirational", + "Male, Deep Voice": "Male, Deep Voice", + "Female, Sweet": "Female, Sweet", + "Male, Cute": "Male, Cute", + "Female, Lovely": "Female, Lovely" } diff --git a/translations/es.json b/translations/es.json index 15fadf5f..ffbc9994 100644 --- a/translations/es.json +++ b/translations/es.json @@ -106,5 +106,20 @@ "No NVIDIA GPU detected": "No se detectó GPU NVIDIA", "No NVIDIA GPU detected or NVIDIA drivers not properly installed": "No se detectó GPU NVIDIA o los controladores NVIDIA no están instalados correctamente", "LLM JSON Format Support": "Soporte de formato JSON para LLM", - "Enable if your LLM supports JSON mode output": "Activar si su LLM admite salida en modo JSON" + "Enable if your LLM supports JSON mode output": "Activar si su LLM admite salida en modo JSON", + "MiniMax API Key": "Clave API de MiniMax", + "MiniMax Voice": "Voz de MiniMax", + "MiniMax TTS Model": "Modelo TTS de MiniMax", + "English Female, Graceful": "Femenina inglesa, elegante", + "English Male, Insightful": "Masculino inglés, perspicaz", + "English Female, Radiant": "Femenina inglesa, radiante", + "English Male, Persuasive": "Masculino inglés, persuasivo", + "English, Robot": "Inglés, robot", + "Female, Wise": "Femenina, sabia", + "Friendly": "Amigable", + "Female, Inspirational": "Femenina, inspiradora", + "Male, Deep Voice": "Masculino, voz profunda", + "Female, Sweet": "Femenina, dulce", + "Male, Cute": "Masculino, adorable", + "Female, Lovely": "Femenina, encantadora" } diff --git a/translations/fr.json b/translations/fr.json index 446f3bbe..eeeba646 100644 --- a/translations/fr.json +++ b/translations/fr.json @@ -106,5 +106,20 @@ "No NVIDIA GPU detected": "Aucun GPU NVIDIA détecté", "No NVIDIA GPU detected or NVIDIA drivers not properly installed": "Aucun GPU NVIDIA détecté ou pilotes NVIDIA mal installés", "LLM JSON Format Support": "Support du format JSON pour LLM", - "Enable if your LLM supports JSON mode output": "Activer si votre LLM prend en charge la sortie en mode JSON" + "Enable if your LLM supports JSON mode output": "Activer si votre LLM prend en charge la sortie en mode JSON", + "MiniMax API Key": "Clé API MiniMax", + "MiniMax Voice": "Voix MiniMax", + "MiniMax TTS Model": "Modèle TTS MiniMax", + "English Female, Graceful": "Femme anglaise, gracieuse", + "English Male, Insightful": "Homme anglais, perspicace", + "English Female, Radiant": "Femme anglaise, radieuse", + "English Male, Persuasive": "Homme anglais, persuasif", + "English, Robot": "Anglais, robot", + "Female, Wise": "Femme, sage", + "Friendly": "Amical", + "Female, Inspirational": "Femme, inspirante", + "Male, Deep Voice": "Homme, voix grave", + "Female, Sweet": "Femme, douce", + "Male, Cute": "Homme, mignon", + "Female, Lovely": "Femme, charmante" } diff --git a/translations/ja.json b/translations/ja.json index 1bce1783..2a16ec4b 100644 --- a/translations/ja.json +++ b/translations/ja.json @@ -106,5 +106,20 @@ "No NVIDIA GPU detected": "NVIDIA GPUが検出されません", "No NVIDIA GPU detected or NVIDIA drivers not properly installed": "NVIDIA GPUが検出されないか、NVIDIAドライバーが正しくインストールされていません", "LLM JSON Format Support": "LLM JSON形式サポート", - "Enable if your LLM supports JSON mode output": "LLMがJSON出力モードをサポートしている場合に有効化" + "Enable if your LLM supports JSON mode output": "LLMがJSON出力モードをサポートしている場合に有効化", + "MiniMax API Key": "MiniMax APIキー", + "MiniMax Voice": "MiniMax 音声", + "MiniMax TTS Model": "MiniMax TTSモデル", + "English Female, Graceful": "英語女性、優雅", + "English Male, Insightful": "英語男性、洞察力", + "English Female, Radiant": "英語女性、明るい", + "English Male, Persuasive": "英語男性、説得力", + "English, Robot": "英語、ロボット", + "Female, Wise": "女性、賢い", + "Friendly": "フレンドリー", + "Female, Inspirational": "女性、インスピレーション", + "Male, Deep Voice": "男性、深い声", + "Female, Sweet": "女性、甘い", + "Male, Cute": "男性、可愛い", + "Female, Lovely": "女性、愛らしい" } diff --git a/translations/ru.json b/translations/ru.json index 07ba21cb..8f02a57c 100644 --- a/translations/ru.json +++ b/translations/ru.json @@ -106,5 +106,20 @@ "No NVIDIA GPU detected": "GPU NVIDIA не обнаружен", "No NVIDIA GPU detected or NVIDIA drivers not properly installed": "GPU NVIDIA не обнаружен или драйверы NVIDIA установлены неправильно", "LLM JSON Format Support": "Поддержка формата JSON для LLM", - "Enable if your LLM supports JSON mode output": "Включите, если ваш LLM поддерживает вывод в формате JSON" + "Enable if your LLM supports JSON mode output": "Включите, если ваш LLM поддерживает вывод в формате JSON", + "MiniMax API Key": "API-ключ MiniMax", + "MiniMax Voice": "Голос MiniMax", + "MiniMax TTS Model": "Модель TTS MiniMax", + "English Female, Graceful": "Английский женский, грациозный", + "English Male, Insightful": "Английский мужской, проницательный", + "English Female, Radiant": "Английский женский, сияющий", + "English Male, Persuasive": "Английский мужской, убедительный", + "English, Robot": "Английский, робот", + "Female, Wise": "Женский, мудрый", + "Friendly": "Дружелюбный", + "Female, Inspirational": "Женский, вдохновляющий", + "Male, Deep Voice": "Мужской, глубокий голос", + "Female, Sweet": "Женский, сладкий", + "Male, Cute": "Мужской, милый", + "Female, Lovely": "Женский, прелестный" } diff --git a/translations/zh-CN.json b/translations/zh-CN.json index 04596db6..2abf5470 100644 --- a/translations/zh-CN.json +++ b/translations/zh-CN.json @@ -106,6 +106,21 @@ "No NVIDIA GPU detected": "未检测到NVIDIA GPU", "No NVIDIA GPU detected or NVIDIA drivers not properly installed": "未检测到NVIDIA GPU或NVIDIA驱动未正确安装", "LLM JSON Format Support": "LLM JSON格式支持", - "Enable if your LLM supports JSON mode output": "如果选用的LLM支持JSON模式输出，请启用" + "Enable if your LLM supports JSON mode output": "如果选用的LLM支持JSON模式输出，请启用", + "MiniMax API Key": "MiniMax API密钥", + "MiniMax Voice": "MiniMax 语音", + "MiniMax TTS Model": "MiniMax TTS模型", + "English Female, Graceful": "英语女声，优雅", + "English Male, Insightful": "英语男声，沉稳", + "English Female, Radiant": "英语女声，活泼", + "English Male, Persuasive": "英语男声，有说服力", + "English, Robot": "英语，机器人", + "Female, Wise": "女声，智慧", + "Friendly": "友善", + "Female, Inspirational": "女声，励志", + "Male, Deep Voice": "男声，低沉", + "Female, Sweet": "女声，甜美", + "Male, Cute": "男声，可爱", + "Female, Lovely": "女声，可爱" } diff --git a/translations/zh-HK.json b/translations/zh-HK.json index cedc73f6..053c859a 100644 --- a/translations/zh-HK.json +++ b/translations/zh-HK.json @@ -106,5 +106,20 @@ "No NVIDIA GPU detected": "未檢測到NVIDIA GPU", "No NVIDIA GPU detected or NVIDIA drivers not properly installed": "未檢測到NVIDIA GPU或NVIDIA驅動未正確安裝", "LLM JSON Format Support": "LLM JSON格式支持", - "Enable if your LLM supports JSON mode output": "如果選用的LLM支持JSON模式輸出，請啟用" + "Enable if your LLM supports JSON mode output": "如果選用的LLM支持JSON模式輸出，請啟用", + "MiniMax API Key": "MiniMax API密鑰", + "MiniMax Voice": "MiniMax 語音", + "MiniMax TTS Model": "MiniMax TTS模型", + "English Female, Graceful": "英語女聲，優雅", + "English Male, Insightful": "英語男聲，沉穩", + "English Female, Radiant": "英語女聲，活潑", + "English Male, Persuasive": "英語男聲，有說服力", + "English, Robot": "英語，機器人", + "Female, Wise": "女聲，智慧", + "Friendly": "友善", + "Female, Inspirational": "女聲，勵志", + "Male, Deep Voice": "男聲，低沉", + "Female, Sweet": "女聲，甜美", + "Male, Cute": "男聲，可愛", + "Female, Lovely": "女聲，可愛" }