From 3130044909fbd61d512286536540da933d9ef9f4 Mon Sep 17 00:00:00 2001
From: Sakura-RanChen <1908198662@qq.com>
Date: Thu, 29 May 2025 09:05:08 +0800
Subject: [PATCH] test

---
 main/xiaozhi-server/config.yaml               |  7 ++
 .../core/providers/tts/linkerai.py            | 95 +++++++++++++++++++
 2 files changed, 102 insertions(+)
 create mode 100644 main/xiaozhi-server/core/providers/tts/linkerai.py

diff --git a/main/xiaozhi-server/config.yaml b/main/xiaozhi-server/config.yaml
index 25c8e017..473c095b 100644
--- a/main/xiaozhi-server/config.yaml
+++ b/main/xiaozhi-server/config.yaml
@@ -713,4 +713,11 @@ TTS:
     headers: # 自定义请求头
       # Authorization: Bearer xxxx
     format: mp3 # 接口返回的音频格式
+    output_dir: tmp/
+  LinkeraiTTS:
+    #各参数意义见开发文档：https://tts.linkerai.top/docs#/default/text_to_speech_tts_get
+    type: linkerai
+    api_url: https://tts.linkerai.top/tts
+    access_token: "test"
+    voice: "OUeAo1mhq6IBExi"
     output_dir: tmp/
\ No newline at end of file
diff --git a/main/xiaozhi-server/core/providers/tts/linkerai.py b/main/xiaozhi-server/core/providers/tts/linkerai.py
new file mode 100644
index 00000000..8ad069bd
--- /dev/null
+++ b/main/xiaozhi-server/core/providers/tts/linkerai.py
@@ -0,0 +1,95 @@
+from core.providers.tts.base import TTSProviderBase
+from core.providers.tts.dto.dto import (
+    TTSMessageDTO,
+    SentenceType,
+    ContentType,
+    InterfaceType
+)
+from config.logger import setup_logging
+from core.utils import opus_encoder_utils
+import requests
+
+TAG = __name__
+logger = setup_logging()
+
+
+class TTSProvider(TTSProviderBase):
+    def __init__(self, config, delete_audio_file):
+        super().__init__(config, delete_audio_file)
+        self.interface_type = InterfaceType.SINGLE_STREAM
+        self.access_token = config.get("access_token")
+        self.voice = config.get("voice")
+        self.api_url = config.get("api_url")
+
+        # 根据配置选择音频格式，优先使用opus
+        self.audio_format = config.get("audio_format", "opus")
+
+        # 创建Opus编码器
+        self.opus_encoder = opus_encoder_utils.OpusEncoderUtils(
+            sample_rate=16000,
+            channels=1,
+            frame_size_ms=60
+        )
+
+    async def text_to_speak(self, text, _):
+        """将文本转换为语音（流式）"""
+        await self.send_text(self.voice, text)
+        return
+
+    async def send_text(self, speaker: str, text: str):
+        """向 TTS 服务发送文本并获取音频流"""
+        try:
+            # 构造请求参数
+            params = {
+                "tts_text": text,
+                "spk_id": self.voice,
+                "frame_duration": 60,
+                "stream": "true",
+                "target_sr": 16000,
+                "audio_format": self.audio_format,
+            }
+
+            # 构造请求头
+            headers = {
+                "Authorization": f"Bearer {self.access_token}",
+            }
+
+            # 发送流式请求
+            response = requests.get(
+                self.api_url,
+                params=params,
+                headers=headers,
+                stream=True
+            )
+
+            # 检查响应状态
+            if response.status_code != 200:
+                logger.bind(tag=TAG).error(f"TTS 请求失败: {response.status_code}, {response.text}")
+                return
+
+            # 处理音频流
+            audio_frames = []
+            for chunk in response.iter_content(chunk_size=1024):
+                if chunk:
+                    if self.audio_format == "pcm":
+                        # 将PCM转换为Opus
+                        opus_frames = self.opus_encoder.encode_pcm_to_opus(chunk, False)
+                        audio_frames.extend(opus_frames)
+                    else:
+                        # 直接使用Opus帧
+                        audio_frames.append(chunk)
+
+            # 将音频帧放入队列
+            self.tts_audio_queue.put(
+                (SentenceType.MIDDLE, audio_frames, text)
+            )
+
+        except Exception as e:
+            logger.bind(tag=TAG).error(f"TTS 流式处理异常：{str(e)}")
+            raise
+
+    async def close(self):
+        """资源清理方法"""
+        await super().close()
+        if hasattr(self, "opus_encoder"):
+            self.opus_encoder.close()