Merge pull request #1253 from hsoftxl/custom_tts_api

自定义tts接口增加POST请求支持

main/manager-api/src/main/resources/db/changelog/202505151450.sql

@@ -0,0 +1,18 @@
+-- 修改自定义TTS接口请求定义
+update `ai_model_provider` set `fields` =
+'[{"key":"url","label":"服务地址","type":"string"},{"key":"method","label":"请求方式","type":"string"},{"key":"params","label":"请求参数","type":"dict","dict_name":"params"},{"key":"headers","label":"请求头","type":"dict","dict_name":"headers"},{"key":"format","label":"音频格式","type":"string"},{"key":"output_dir","label":"输出目录","type":"string"}]'
+where `id` = 'SYSTEM_TTS_custom';
+
+-- 修改自定义TTS配置说明
+UPDATE `ai_model_config` SET
+`doc_link` = NULL,
+`remark` = '自定义TTS配置说明：
+1. 支持自定义TTS接口服务
+2. 使用GET/POST方式请求
+3. 需要网络连接
+4. 输出文件保存在tmp/目录
+配置说明：
+1. 在params中配置请求参数,使用JSON格式
+   例如：{"text": "{prompt_text}", "voice": "{voice}", "speed": "{speed}"}
+2. 在headers中配置请求头
+3. 设置返回音频格式' WHERE `id` = 'TTS_CustomTTS';

main/manager-api/src/main/resources/db/changelog/db.changelog-master.yaml

@@ -156,3 +156,10 @@ databaseChangeLog:
         - sqlFile:
             encoding: utf8
             path: classpath:db/changelog/202505201744.sql
+  - changeSet:
+      id: 202505151450
+      author: hsoftxl
+      changes:
+        - sqlFile:
+            encoding: utf8
+            path: classpath:db/changelog/202505151450.sql

main/xiaozhi-server/core/providers/tts/custom.py

@@ -1,4 +1,5 @@
 import os
+import json
 import uuid
 import requests
 from config.logger import setup_logging
@@ -12,11 +13,21 @@ class TTSProvider(TTSProviderBase):
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
         self.url = config.get("url")
+        self.method = config.get("method", "GET")
         self.headers = config.get("headers", {})
-        self.params = config.get("params")
         self.format = config.get("format", "wav")
         self.output_file = config.get("output_dir", "tmp/")
 
+        self.params = config.get("params")
+
+        if isinstance(self.params, str):
+            try:
+                self.params = json.loads(self.params)
+            except json.JSONDecodeError:
+                raise ValueError("Custom TTS配置参数出错,无法将字符串解析为对象")
+        elif not isinstance(self.params, dict):
+            raise TypeError("Custom TTS配置参数出错, 请参考配置说明")
+
     def generate_filename(self):
         return os.path.join(self.output_file, f"tts-{datetime.now().date()}@{uuid.uuid4().hex}.{self.format}")
 
@@ -27,7 +38,10 @@ class TTSProvider(TTSProviderBase):
                 v = v.replace("{prompt_text}", text)
             request_params[k] = v
 
-        resp = requests.get(self.url, params=request_params, headers=self.headers)
+        if self.method.upper() == "POST":
+            resp = requests.post(self.url, json=request_params, headers=self.headers)
+        else:
+            resp = requests.get(self.url, params=request_params, headers=self.headers)
         if resp.status_code == 200:
             with open(output_file, "wb") as file:
                 file.write(resp.content)