add edge tts

2025-11-28 20:27:10 +08:00
parent f796a3833b
commit 87160c5265
20 changed files with 3589 additions and 3 deletions
--- a/tts/service.py
+++ b/tts/service.py
@ -0,0 +1,119 @@
+"""
+TTS 服务集成模块
+
+提供高层 TTS 服务接口，直接使用配置文件中的 TTS 设置。
+"""
+
+from io import BytesIO
+from typing import Optional
+from config.settings import settings
+from tts.factory import TTSEngineFactory
+from tts.base import TTSEngine
+from utils.logger import logger
+
+
+class TTSService:
+    """
+    TTS 服务
+    
+    提供统一的 TTS 调用接口，自动使用配置文件中的引擎和参数。
+    """
+
+    _engine: Optional[TTSEngine] = None
+
+    @classmethod
+    def _get_engine(cls) -> TTSEngine:
+        """
+        获取 TTS 引擎实例
+
+        Returns:
+            TTSEngine 实例
+        """
+        if cls._engine is None:
+            cls._engine = TTSEngineFactory.create(settings.TTS_ENGINE)
+            logger.info(
+                f"TTS Service initialized with engine: {settings.TTS_ENGINE}"
+            )
+        return cls._engine
+
+    @classmethod
+    async def synthesize(
+        cls,
+        text: str,
+        language: Optional[str] = None,
+        voice: Optional[str] = None,
+        rate: Optional[float] = None,
+        pitch: Optional[float] = None,
+    ) -> BytesIO:
+        """
+        将文本合成为语音
+
+        Args:
+            text: 要合成的文本
+            language: 语言代码，默认使用配置文件中的 TTS_LANGUAGE
+            voice: 声音 ID，默认使用配置文件中的 TTS_VOICE
+            rate: 语速，默认使用配置文件中的 TTS_RATE
+            pitch: 音调，默认使用配置文件中的 TTS_PITCH
+
+        Returns:
+            BytesIO 对象，包含合成后的音频数据
+        """
+        engine = cls._get_engine()
+
+        # 使用配置文件中的默认值
+        language = language or settings.TTS_LANGUAGE
+        voice = voice or settings.TTS_VOICE or None
+        rate = rate or settings.TTS_RATE
+        pitch = pitch or settings.TTS_PITCH
+
+        logger.debug(f"Synthesizing text with TTS Service: {text[:50]}...")
+
+        return await engine.synthesize(
+            text=text,
+            language=language,
+            voice=voice,
+            rate=rate,
+            pitch=pitch,
+        )
+
+    @classmethod
+    async def get_supported_voices(cls, language: Optional[str] = None) -> list[dict]:
+        """
+        获取支持的声音列表
+
+        Args:
+            language: 语言代码，默认使用配置文件中的 TTS_LANGUAGE
+
+        Returns:
+            声音列表
+        """
+        engine = cls._get_engine()
+        language = language or settings.TTS_LANGUAGE
+        return await engine.get_supported_voices(language)
+
+    @classmethod
+    def get_engine_info(cls) -> dict:
+        """
+        获取当前 TTS 引擎信息
+
+        Returns:
+            包含引擎名称、版本、当前配置等信息的字典
+        """
+        engine = cls._get_engine()
+        return {
+            "engine_name": engine.get_engine_name(),
+            "engine_version": engine.get_engine_version(),
+            "config": {
+                "language": settings.TTS_LANGUAGE,
+                "voice": settings.TTS_VOICE or "default",
+                "rate": settings.TTS_RATE,
+                "pitch": settings.TTS_PITCH,
+            },
+        }
+
+    @classmethod
+    def reset_engine(cls) -> None:
+        """重置 TTS 引擎实例（仅在切换引擎时需要调用）"""
+        cls._engine = None
+        TTSEngineFactory.clear_instances()
+        logger.info("TTS Service engine reset")