Merge pull request #1551 from GowayLee/master

Feature: 添加对 MiniMax TTS API的支持

astrbot/core/config/default.py

@@ -820,6 +820,27 @@ CONFIG_METADATA_2 = {
                         "azure_tts_subscription_key": "",
                         "azure_tts_region": "eastus"
                     },
+                    "MiniMax TTS(API)": {
+                        "id": "minimax_tts",
+                        "type": "minimax_tts_api",
+                        "provider_type": "text_to_speech",
+                        "enable": False,
+                        "api_key": "",
+                        "api_base": "https://api.minimax.chat/v1/t2a_v2",
+                        "minimax-group-id": "",
+                        "model": "speech-02-turbo",
+                        "minimax-langboost": "auto",
+                        "minimax-voice-speed": 1.0,
+                        "minimax-voice-vol": 1.0,
+                        "minimax-voice-pitch": 0,
+                        "minimax-is-timber-weight": False,
+                        "minimax-voice-id": "female-shaonv",
+                        "minimax-timber-weight": '[\n    {\n        "voice_id": "Chinese (Mandarin)_Warm_Girl",\n        "weight": 25\n    },\n    {\n        "voice_id": "Chinese (Mandarin)_BashfulGirl",\n        "weight": 50\n    }\n]',
+                        "minimax-voice-emotion": "neutral",
+                        "minimax-voice-latex": False,
+                        "minimax-voice-english-normalization": False,
+                        "timeout": 20,
+                    },
                 },
                 "items": {
                     "azure_tts_voice": {
@@ -943,6 +964,64 @@ CONFIG_METADATA_2 = {
                             },
                         },
                     },
+                    "minimax-group-id": {
+                        "type": "string",
+                        "description": "用户组",
+                        "hint": "于账户管理->基本信息中可见",
+                    },
+                    "minimax-langboost": {
+                        "type": "string",
+                        "description": "指定语言/方言",
+                        "hint": "增强对指定的小语种和方言的识别能力，设置后可以提升在指定小语种/方言场景下的语音表现",
+                        "options": [ "Chinese","Chinese,Yue","English","Arabic","Russian","Spanish","French","Portuguese","German","Turkish","Dutch","Ukrainian","Vietnamese","Indonesian","Japanese","Italian","Korean","Thai","Polish","Romanian","Greek","Czech","Finnish","Hindi","auto",],
+                    },
+                    "minimax-voice-speed": {
+                        "type": "float",
+                        "description": "语速",
+                        "hint": "生成声音的语速, 取值[0.5, 2], 默认为1.0, 取值越大，语速越快",
+                    },
+                    "minimax-voice-vol": {
+                        "type": "float",
+                        "description": "音量",
+                        "hint": "生成声音的音量, 取值(0, 10], 默认为1.0, 取值越大，音量越高",
+                    },
+                    "minimax-voice-pitch": {
+                        "type": "int",
+                        "description": "语调",
+                        "hint": "生成声音的语调, 取值[-12, 12], 默认为0",
+                    },
+                    "minimax-is-timber-weight": {
+                        "type": "bool",
+                        "description": "启用混合音色",
+                        "hint": "启用混合音色, 支持以自定义权重混合最多四种音色, 启用后自动忽略单一音色设置",
+                    },
+                    "minimax-timber-weight": {
+                        "type": "string",
+                        "description": "混合音色",
+                        "editor_mode": True,
+                        "hint": "混合音色及其权重, 最多支持四种音色, 权重为整数, 取值[1, 100]. 可在官网API语音调试台预览代码获得预设以及编写模板, 需要严格按照json字符串格式编写, 可以查看控制台判断是否解析成功. 具体结构可参照默认值以及官网代码预览.",
+                    },
+                    "minimax-voice-id": {
+                        "type": "string",
+                        "description": "单一音色",
+                        "hint": "单一音色编号, 详见官网文档",
+                    },
+                    "minimax-voice-emotion": {
+                        "type": "string",
+                        "description": "情绪",
+                        "hint": "控制合成语音的情绪",
+                        "options": ["happy","sad","angry","fearful","disgusted","surprised","neutral",],
+                    },
+                    "minimax-voice-latex": {
+                        "type": "bool",
+                        "description": "支持朗读latex公式",
+                        "hint": "朗读latex公式, 但是需要确保输入文本按官网要求格式化",
+                    },
+                    "minimax-voice-english-normalization": {
+                        "type": "bool",
+                        "description": "支持英语文本规范化",
+                        "hint": "可提升数字阅读场景的性能，但会略微增加延迟",
+                    },
                     "rag_options": {
                         "description": "RAG 选项",
                         "type": "object",

astrbot/core/provider/manager.py

@@ -206,6 +206,10 @@ class ProviderManager:
                     from .sources.azure_tts_source import (
                         AzureTTSProvider as AzureTTSProvider,
                     )
+                case "minimax_tts_api":
+                    from .sources.minimax_tts_api_source import (
+                        ProviderMiniMaxTTSAPI as ProviderMiniMaxTTSAPI,
+                    )
         except (ImportError, ModuleNotFoundError) as e:
             logger.critical(
                 f"加载 {provider_config['type']}({provider_config['id']}) 提供商适配器失败：{e}。可能是因为有未安装的依赖。"

astrbot/core/provider/sources/minimax_tts_api_source.py

@@ -0,0 +1,149 @@
+import json
+import os
+import uuid
+import aiohttp
+from typing import Dict, List, Union, AsyncIterator
+from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.api import logger
+from ..entities import ProviderType
+from ..provider import TTSProvider
+from ..register import register_provider_adapter
+
+
+@register_provider_adapter(
+    "minimax_tts_api", "MiniMax TTS API", provider_type=ProviderType.TEXT_TO_SPEECH
+)
+class ProviderMiniMaxTTSAPI(TTSProvider):
+    def __init__(
+        self,
+        provider_config: dict,
+        provider_settings: dict,
+    ) -> None:
+        super().__init__(provider_config, provider_settings)
+        self.chosen_api_key: str = provider_config.get("api_key", "")
+        self.api_base: str = provider_config.get(
+            "api_base", "https://api.minimax.chat/v1/t2a_v2"
+        )
+        self.group_id: str = provider_config.get("minimax-group-id", "")
+        self.set_model(provider_config.get("model", ""))
+        self.lang_boost: str = provider_config.get("minimax-langboost", "auto")
+        self.is_timber_weight: bool = provider_config.get(
+            "minimax-is-timber-weight", False
+        )
+        self.timber_weight: List[Dict[str, Union[str, int]]] = json.loads(
+            provider_config.get(
+                "minimax-timber-weight",
+                '[{"voice_id": "Chinese (Mandarin)_Warm_Girl", "weight": 1}]',
+            )
+        )
+
+        self.voice_setting: dict = {
+            "speed": provider_config.get("minimax-voice-speed", 1.0),
+            "vol": provider_config.get("minimax-voice-vol", 1.0),
+            "pitch": provider_config.get("minimax-voice-pitch", 0),
+            "voice_id": ""
+            if self.is_timber_weight
+            else provider_config.get("minimax-voice-id", ""),
+            "emotion": provider_config.get("minimax-voice-emotion", "neutral"),
+            "latex_read": provider_config.get("minimax-voice-latex", False),
+            "english_normalization": provider_config.get(
+                "minimax-voice-english-normalization", False
+            ),
+        }
+
+        self.audio_setting: dict = {
+            "sample_rate": 32000,
+            "bitrate": 128000,
+            "format": "mp3",
+        }
+
+        self.concat_base_url: str = f"{self.api_base}?GroupId={self.group_id}"
+        self.headers = {
+            "Authorization": f"Bearer {self.chosen_api_key}",
+            "accept": "application/json, text/plain, */*",
+            "content-type": "application/json",
+        }
+
+    def _build_tts_stream_body(self, text: str):
+        """构建流式请求体"""
+        dict_body: Dict[str, object] = {
+            "model": self.model_name,
+            "text": text,
+            "stream": True,
+            "language_boost": self.lang_boost,
+            "voice_setting": self.voice_setting,
+            "audio_setting": self.audio_setting,
+        }
+        if self.is_timber_weight:
+            dict_body["timber_weights"] = self.timber_weight
+
+        return json.dumps(dict_body)
+
+    async def _call_tts_stream(self, text: str) -> AsyncIterator[bytes]:
+        """进行流式请求"""
+        try:
+            async with aiohttp.ClientSession() as session:
+                async with session.post(
+                    self.concat_base_url,
+                    headers=self.headers,
+                    data=self._build_tts_stream_body(text),
+                    timeout=aiohttp.ClientTimeout(total=60),
+                ) as response:
+                    response.raise_for_status()
+
+                    buffer = b""
+                    while True:
+                        chunk = await response.content.read(8192)
+                        if not chunk:
+                            break
+
+                        buffer += chunk
+
+                        while b"\n\n" in buffer:
+                            try:
+                                message, buffer = buffer.split(b"\n\n", 1)
+                                if message.startswith(b"data: "):
+                                    try:
+                                        data = json.loads(message[6:])
+                                        if "extra_info" in data:
+                                            continue
+                                        audio = data.get("data", {}).get("audio")
+                                        if audio is not None:
+                                            yield audio
+                                    except json.JSONDecodeError:
+                                        logger.warning(
+                                            "Failed to parse JSON data from SSE message"
+                                        )
+                                        continue
+                            except ValueError:
+                                buffer = buffer[-1024:]
+
+        except aiohttp.ClientError as e:
+            raise Exception(f"MiniMax TTS API请求失败: {str(e)}")
+
+    async def _audio_play(self, audio_stream: AsyncIterator[str]) -> bytes:
+        """解码数据流到 audio 比特流"""
+        chunks = []
+        async for chunk in audio_stream:
+            if chunk.strip():
+                chunks.append(bytes.fromhex(chunk.strip()))
+        return b"".join(chunks)
+
+    async def get_audio(self, text: str) -> str:
+        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        os.makedirs(temp_dir, exist_ok=True)
+        path = os.path.join(temp_dir, f"minimax_tts_api_{uuid.uuid4()}.mp3")
+
+        try:
+            # 直接将异步生成器传递给 _audio_play 方法
+            audio_stream = self._call_tts_stream(text)
+            audio = await self._audio_play(audio_stream)
+
+            # 结果保存至文件
+            with open(path, "wb") as file:
+                file.write(audio)
+
+            return path
+
+        except aiohttp.ClientError as e:
+            raise e