feat: add xiaomi MiMo TTS & STT providers

astrbot/core/config/default.py

@@ -1461,6 +1461,23 @@ class ChatProviderTemplate(TypedDict):
                         "timeout": "20",
                         "proxy": "",
                     },
+                    "MiMo TTS(API)": {
+                        "id": "mimo_tts",
+                        "type": "mimo_tts_api",
+                        "provider": "mimo",
+                        "provider_type": "text_to_speech",
+                        "enable": False,
+                        "api_key": "",
+                        "api_base": "https://api.xiaomimimo.com/v1",
+                        "model": "mimo-v2-tts",
+                        "mimo-tts-voice": "mimo_default",
+                        "mimo-tts-format": "wav",
+                        "mimo-tts-style-prompt": "",
+                        "mimo-tts-dialect": "",
+                        "mimo-tts-seed-text": "Hello, MiMo, have you had lunch?",
+                        "timeout": "20",
+                        "proxy": "",
+                    },
                     "Genie TTS": {
                         "id": "genie_tts",
                         "provider": "genie_tts",
@@ -2314,6 +2331,31 @@ class ChatProviderTemplate(TypedDict):
                         "type": "string",
                         "hint": "OpenAI TTS 的声音。OpenAI 默认支持：'alloy', 'echo', 'fable', 'onyx', 'nova', 'shimmer'",
                     },
+                    "mimo-tts-voice": {
+                        "description": "音色",
+                        "type": "string",
+                        "hint": "MiMo TTS 的音色名称。默认值为 'mimo_default'。",
+                    },
+                    "mimo-tts-format": {
+                        "description": "输出格式",
+                        "type": "string",
+                        "hint": "MiMo TTS 生成音频的格式，例如 'wav'。",
+                    },
+                    "mimo-tts-style-prompt": {
+                        "description": "风格提示词",
+                        "type": "string",
+                        "hint": "用于控制生成语音的说话风格、语气或情绪，例如温柔、活泼、沉稳等。可留空。",
+                    },
+                    "mimo-tts-dialect": {
+                        "description": "方言",
+                        "type": "string",
+                        "hint": "指定生成语音时使用的方言或口音，例如四川话、粤语口音等。可留空。",
+                    },
+                    "mimo-tts-seed-text": {
+                        "description": "种子文本",
+                        "type": "string",
+                        "hint": "用于引导音色和说话方式的参考文本，会影响生成语音的表达风格。",
+                    },
                     "fishaudio-tts-character": {
                         "description": "character",
                         "type": "string",

astrbot/core/provider/manager.py

@@ -399,6 +399,10 @@ def dynamic_import_provider(self, type: str) -> None:
                 from .sources.openai_tts_api_source import (
                     ProviderOpenAITTSAPI as ProviderOpenAITTSAPI,
                 )
+            case "mimo_tts_api":
+                from .sources.mimo_tts_api_source import (
+                    ProviderMiMoTTSAPI as ProviderMiMoTTSAPI,
+                )
             case "genie_tts":
                 from .sources.genie_tts import (
                     GenieTTSProvider as GenieTTSProvider,

astrbot/core/provider/sources/mimo_tts_api_source.py

@@ -0,0 +1,149 @@
+import base64
+import uuid
+from pathlib import Path
+
+import httpx
+
+from ..entities import ProviderType
+from ..provider import TTSProvider
+from ..register import register_provider_adapter
+from astrbot import logger
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
+
+
+def normalize_timeout(timeout: int | str | None) -> int | None:
+    if timeout in (None, ""):
+        return None
+    if isinstance(timeout, str):
+        return int(timeout)
+    return timeout
+
+
+def build_headers(api_key: str) -> dict[str, str]:
+    headers = {"Content-Type": "application/json"}
+    if api_key:
+        headers["api-key"] = api_key
+        headers["Authorization"] = f"Bearer {api_key}"
+    return headers
+
+
+def get_temp_dir() -> Path:
+    temp_dir = Path(get_astrbot_temp_path())
+    temp_dir.mkdir(parents=True, exist_ok=True)
+    return temp_dir
+
+
+def create_http_client(timeout: int | None, proxy: str) -> httpx.AsyncClient:
+    client_kwargs: dict[str, object] = {
+        "timeout": timeout,
+        "follow_redirects": True,
+    }
+    if proxy:
+        logger.info("[MiMo API] Using proxy: %s", proxy)
+        client_kwargs["proxy"] = proxy
+    return httpx.AsyncClient(**client_kwargs)
+
+
+def build_api_url(api_base: str) -> str:
+    normalized_api_base = api_base.rstrip("/")
+    if normalized_api_base.endswith("/chat/completions"):
+        return normalized_api_base
+    return normalized_api_base + "/chat/completions"
+
+
+@register_provider_adapter(
+    "mimo_tts_api",
+    "MiMo TTS API",
+    provider_type=ProviderType.TEXT_TO_SPEECH,
+)
+class ProviderMiMoTTSAPI(TTSProvider):
+    def __init__(
+        self,
+        provider_config: dict,
+        provider_settings: dict,
+    ) -> None:
+        super().__init__(provider_config, provider_settings)
+        self.chosen_api_key = provider_config.get("api_key", "")
+        self.api_base = provider_config.get(
+            "api_base",
+            "https://api.xiaomimimo.com/v1",
+        )
+        self.proxy = provider_config.get("proxy", "")
+        self.timeout = normalize_timeout(provider_config.get("timeout", 20))
+        self.voice = provider_config.get("mimo-tts-voice", "mimo_default")
+        self.audio_format = provider_config.get("mimo-tts-format", "wav")
+        self.style_prompt = provider_config.get("mimo-tts-style-prompt", "")
+        self.dialect = provider_config.get("mimo-tts-dialect", "")
+        self.seed_text = provider_config.get(
+            "mimo-tts-seed-text",
+            "Hello, MiMo, have you had lunch?",
+        )
+        self.set_model(provider_config.get("model", "mimo-v2-tts"))
+        self.client = create_http_client(self.timeout, self.proxy)
+
+    def _build_user_prompt(self) -> str:
+        prompt_parts: list[str] = []
+
+        if self.style_prompt.strip():
+            prompt_parts.append(self.style_prompt.strip())
+        if self.dialect.strip():
+            prompt_parts.append(f"Please use {self.dialect.strip()} when speaking.")
+
+        if not prompt_parts:
+            return self.seed_text
+
+        if self.seed_text.strip():
+            prompt_parts.append(self.seed_text.strip())
+
+        return " ".join(prompt_parts)
+
+    def _build_payload(self, text: str) -> dict:
+        return {
+            "model": self.model_name,
+            "messages": [
+                {
+                    "role": "user",
+                    "content": self._build_user_prompt(),
+                },
+                {
+                    "role": "assistant",
+                    "content": text,
+                },
+            ],
+            "audio": {
+                "format": self.audio_format,
+                "voice": self.voice,
+            },
+        }
+
+    async def get_audio(self, text: str) -> str:
+        response = await self.client.post(
+            build_api_url(self.api_base),
+            headers=build_headers(self.chosen_api_key),
+            json=self._build_payload(text),
+        )
+
+        try:
+            response.raise_for_status()
+        except Exception as exc:
+            error_text = response.text[:1024]
+            raise Exception(
+                f"MiMo TTS API request failed: HTTP {response.status_code}, response: {error_text}"
+            ) from exc
+
+        data = response.json()
+        audio_data = (
+            data.get("choices", [{}])[0].get("message", {}).get("audio", {}).get("data")
+        )
+        if not audio_data:
+            raise Exception(f"MiMo TTS API returned no audio payload: {data}")
+
+        output_path = (
+            get_temp_dir() / f"mimo_tts_api_{uuid.uuid4()}.{self.audio_format}"
+        )
+        output_path.write_bytes(base64.b64decode(audio_data))
+        return str(output_path)
+
+    async def terminate(self):
+        if self.client:
+            await self.client.aclose()

dashboard/src/composables/useProviderSources.ts

@@ -325,6 +325,7 @@ export function useProviderSources(options: UseProviderSourcesOptions) {
       openai_whisper_selfhost: 'speech_to_text',
       sensevoice_stt_selfhost: 'speech_to_text',
       openai_tts_api: 'text_to_speech',
+      mimo_tts_api: 'text_to_speech',
       edge_tts: 'text_to_speech',
       gsvi_tts_api: 'text_to_speech',
       fishaudio_tts_api: 'text_to_speech',

dashboard/src/i18n/locales/en-US/features/config-metadata.json

@@ -1391,6 +1391,26 @@
         "description": "voice",
         "hint": "OpenAI TTS voice. OpenAI defaults: 'alloy', 'echo', 'fable', 'onyx', 'nova', 'shimmer'."
       },
+      "mimo-tts-voice": {
+        "description": "Voice",
+        "hint": "MiMo TTS voice name. Default is 'mimo_default'."
+      },
+      "mimo-tts-format": {
+        "description": "Output format",
+        "hint": "Audio format generated by MiMo TTS, for example 'wav'."
+      },
+      "mimo-tts-style-prompt": {
+        "description": "Style prompt",
+        "hint": "Guides speaking style, tone, or emotion such as gentle, lively, or calm. Optional."
+      },
+      "mimo-tts-dialect": {
+        "description": "Dialect",
+        "hint": "Target dialect or accent for generated speech, such as Sichuan dialect. Optional."
+      },
+      "mimo-tts-seed-text": {
+        "description": "Seed text",
+        "hint": "Reference text used to guide voice characteristics and speaking style."
+      },
       "fishaudio-tts-character": {
         "description": "character",
         "hint": "Fishaudio TTS character. Default is Klee. More roles: https://fish.audio/zh-CN/discovery"
@@ -1518,4 +1538,4 @@
     "helpMiddle": "or",
     "helpSuffix": "."
   }
-}
+}

dashboard/src/i18n/locales/ru-RU/features/config-metadata.json

@@ -1396,6 +1396,26 @@
                 "description": "API Base URL",
                 "hint": "Голоса OpenAI TTS: alloy, echo и др."
             },
+            "mimo-tts-voice": {
+                "description": "Голос",
+                "hint": "Имя голоса MiMo TTS. По умолчанию 'mimo_default'."
+            },
+            "mimo-tts-format": {
+                "description": "Формат вывода",
+                "hint": "Формат аудио, создаваемого MiMo TTS, например 'wav'."
+            },
+            "mimo-tts-style-prompt": {
+                "description": "Подсказка стиля",
+                "hint": "Задает стиль речи, тон или эмоцию, например мягкий, живой или спокойный. Необязательно."
+            },
+            "mimo-tts-dialect": {
+                "description": "Диалект",
+                "hint": "Диалект или акцент для синтезируемой речи, например сычуаньский диалект. Необязательно."
+            },
+            "mimo-tts-seed-text": {
+                "description": "Начальный текст",
+                "hint": "Эталонный текст, который помогает задать особенности голоса и манеру речи."
+            },
             "fishaudio-tts-character": {
                 "description": "Персонаж",
                 "hint": "Персонаж Fishaudio. По умолчанию Klee."
@@ -1523,4 +1543,4 @@
         "helpMiddle": "или",
         "helpSuffix": "."
     }
-}
+}

dashboard/src/i18n/locales/zh-CN/features/config-metadata.json

@@ -1393,6 +1393,26 @@
         "description": "voice",
         "hint": "OpenAI TTS 的声音。OpenAI 默认支持：'alloy', 'echo', 'fable', 'onyx', 'nova', 'shimmer'"
       },
+      "mimo-tts-voice": {
+        "description": "音色",
+        "hint": "MiMo TTS 的音色名称。默认值为 'mimo_default'。"
+      },
+      "mimo-tts-format": {
+        "description": "输出格式",
+        "hint": "MiMo TTS 生成音频的格式，例如 'wav'。"
+      },
+      "mimo-tts-style-prompt": {
+        "description": "风格提示词",
+        "hint": "用于控制生成语音的说话风格、语气或情绪，例如温柔、活泼、沉稳等。可留空。"
+      },
+      "mimo-tts-dialect": {
+        "description": "方言",
+        "hint": "指定生成语音时使用的方言或口音，例如四川话、粤语口音等。可留空。"
+      },
+      "mimo-tts-seed-text": {
+        "description": "种子文本",
+        "hint": "用于引导音色和说话方式的参考文本，会影响生成语音的表达风格。"
+      },
       "fishaudio-tts-character": {
         "description": "character",
         "hint": "fishaudio TTS 的角色。默认为可莉。更多角色请访问：https://fish.audio/zh-CN/discovery"
@@ -1520,4 +1540,4 @@
     "helpMiddle": "或",
     "helpSuffix": "。"
   }
-}
+}

dashboard/src/utils/providerUtils.js

@@ -30,6 +30,7 @@ export function getProviderIcon(type) {
     'lm_studio': 'https://cdn.jsdelivr.net/npm/@lobehub/icons-static-svg@latest/icons/lmstudio.svg',
     'fishaudio': 'https://cdn.jsdelivr.net/npm/@lobehub/icons-static-svg@latest/icons/fishaudio.svg',
     'minimax': 'https://cdn.jsdelivr.net/npm/@lobehub/icons-static-svg@latest/icons/minimax.svg',
+    'mimo': 'https://platform.xiaomimimo.com/favicon.874c9507.png',
     '302ai': 'https://cdn.jsdelivr.net/npm/@lobehub/icons-static-svg@1.53.0/icons/ai302-color.svg',
     'microsoft': 'https://cdn.jsdelivr.net/npm/@lobehub/icons-static-svg@latest/icons/microsoft.svg',
     'vllm': 'https://cdn.jsdelivr.net/npm/@lobehub/icons-static-svg@latest/icons/vllm.svg',