Merge branch 'main' into agent-plugin

main/manager-api/src/main/java/xiaozhi/modules/agent/controller/AgentController.java

@@ -110,6 +110,7 @@ public class AgentController {
             entity.setAsrModelId(template.getAsrModelId());
             entity.setVadModelId(template.getVadModelId());
             entity.setLlmModelId(template.getLlmModelId());
+            entity.setVllmModelId(template.getVllmModelId());
             entity.setTtsModelId(template.getTtsModelId());
             entity.setTtsVoiceId(template.getTtsVoiceId());
             entity.setMemModelId(template.getMemModelId());

main/manager-web/src/views/DictManagement.vue

@@ -538,7 +538,7 @@ export default {
 
 .data-table {
     border-radius: 6px;
-    overflow-y: auto;
+    overflow-y: hidden;
     background-color: transparent !important;
     --table-max-height: calc(100vh - 40vh);
     max-height: var(--table-max-height);

main/xiaozhi-server/core/handle/helloHandle.py

@@ -2,6 +2,7 @@ import time
 import json
 import random
 import asyncio
+from core.utils.dialogue import Message
 from core.utils.util import audio_to_data
 from core.handle.sendAudioHandle import sendAudioMessage, send_stt_message
 from core.utils.util import remove_punctuation_and_length, opus_datas_to_wav_bytes
@@ -67,20 +68,32 @@ async def checkWakeupWords(conn, text):
 
     # 获取当前音色
     voice = getattr(conn.tts, "voice", "default")
+    if not voice:
+        voice = "default"
 
     # 获取唤醒词回复配置
     response = wakeup_words_config.get_wakeup_response(voice)
+    if not response or not response.get("file_path"):
+        response = {
+            "voice": "default",
+            "file_path": "config/assets/wakeup_words.wav",
+            "time": 0,
+            "text": "哈啰啊，我是小智啦，声音好听的台湾女孩一枚，超开心认识你耶，最近在忙啥，别忘了给我来点有趣的料哦，我超爱听八卦的啦",
+        }
 
     # 播放唤醒词回复
     conn.client_abort = False
-    opus_packets, _ = audio_to_data(response["file_path"])
+    opus_packets, _ = audio_to_data(response.get("file_path"))
 
-    conn.logger.bind(tag=TAG).info(f"播放唤醒词回复: {response['text']}")
-    await sendAudioMessage(conn, SentenceType.FIRST, opus_packets, response["text"])
+    conn.logger.bind(tag=TAG).info(f"播放唤醒词回复: {response.get('text')}")
+    await sendAudioMessage(conn, SentenceType.FIRST, opus_packets, response.get("text"))
     await sendAudioMessage(conn, SentenceType.LAST, [], None)
 
+    # 补充对话
+    conn.dialogue.put(Message(role="assistant", content=response.get("text")))
+
     # 检查是否需要更新唤醒词回复
-    if time.time() - response["time"] > WAKEUP_CONFIG["refresh_time"]:
+    if time.time() - response.get("time", 0) > WAKEUP_CONFIG["refresh_time"]:
         if not _wakeup_response_lock.locked():
             asyncio.create_task(wakeupWordsResponse(conn))
     return True
@@ -100,7 +113,7 @@ async def wakeupWordsResponse(conn):
         question = (
             "此刻用户正在和你说```"
             + wakeup_word
-            + "```。\n请你根据以上用户的内容进行简短回复。要像一个人正常人一样说话，不要像机器人一样说话。\n"
+            + "```。\n请你根据以上用户的内容进行20-30字回复。要符合系统设置的角色情感和态度，不要像机器人一样说话。\n"
             + "请勿对这条内容本身进行任何解释和回应，请勿返回表情符号，仅返回对用户的内容的回复。"
         )
 

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

@@ -1,4 +1,5 @@
 import os
+import re
 import queue
 import uuid
 import asyncio
@@ -169,15 +170,18 @@ class TTSProviderBase(ABC):
                 content_type=ContentType.ACTION,
             )
         )
-        self.tts_text_queue.put(
-            TTSMessageDTO(
-                sentence_id=sentence_id,
-                sentence_type=SentenceType.MIDDLE,
-                content_type=content_type,
-                content_detail=content_detail,
-                content_file=content_file,
+        # 对于单句的文本，进行分段处理
+        segments = re.split(r'([。！？!?；;\n])', content_detail)
+        for seg in segments:
+            self.tts_text_queue.put(
+                TTSMessageDTO(
+                    sentence_id=sentence_id,
+                    sentence_type=SentenceType.MIDDLE,
+                    content_type=content_type,
+                    content_detail=seg,
+                    content_file=content_file,
+                )
             )
-        )
         self.tts_text_queue.put(
             TTSMessageDTO(
                 sentence_id=sentence_id,

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

@@ -11,8 +11,8 @@ class TTSProvider(TTSProviderBase):
             self.voice = config.get("private_voice")
         else:
             self.voice = config.get("voice")
-        self.response_format = config.get("response_format", "mp3")
-        self.audio_file_type = config.get("response_format", "mp3")
+        self.response_format = config.get("response_format", "wav")
+        self.audio_file_type = config.get("response_format", "wav")
         self.host = "api.coze.cn"
         self.api_url = f"https://{self.host}/v1/audio/speech"
 

main/xiaozhi-server/core/utils/textUtils.py

@@ -9,26 +9,38 @@ def get_string_no_punctuation_or_emoji(s):
     end = len(chars) - 1
     while end >= start and is_punctuation_or_emoji(chars[end]):
         end -= 1
-    return ''.join(chars[start:end + 1])
+    return "".join(chars[start : end + 1])
+
 
 def is_punctuation_or_emoji(char):
     """检查字符是否为空格、指定标点或表情符号"""
     # 定义需要去除的中英文标点（包括全角/半角）
     punctuation_set = {
-        '，', ',',  # 中文逗号 + 英文逗号
-        '。', '.',  # 中文句号 + 英文句号
-        '！', '!',  # 中文感叹号 + 英文感叹号
-        '-', '－',  # 英文连字符 + 中文全角横线
-        '、'  # 中文顿号
+        "，",
+        ",",  # 中文逗号 + 英文逗号
+        "。",
+        ".",  # 中文句号 + 英文句号
+        "！",
+        "!",  # 中文感叹号 + 英文感叹号
+        "-",
+        "－",  # 英文连字符 + 中文全角横线
+        "、",  # 中文顿号
+        "[",
+        "]",  # 方括号
+        "【",
+        "】",  # 中文方括号
     }
     if char.isspace() or char in punctuation_set:
         return True
     # 检查表情符号（保留原有逻辑）
     code_point = ord(char)
     emoji_ranges = [
-        (0x1F600, 0x1F64F), (0x1F300, 0x1F5FF),
-        (0x1F680, 0x1F6FF), (0x1F900, 0x1F9FF),
-        (0x1FA70, 0x1FAFF), (0x2600, 0x26FF),
-        (0x2700, 0x27BF)
+        (0x1F600, 0x1F64F),
+        (0x1F300, 0x1F5FF),
+        (0x1F680, 0x1F6FF),
+        (0x1F900, 0x1F9FF),
+        (0x1FA70, 0x1FAFF),
+        (0x2600, 0x26FF),
+        (0x2700, 0x27BF),
     ]
-    return any(start <= code_point <= end for start, end in emoji_ranges)
+    return any(start <= code_point <= end for start, end in emoji_ranges)

main/xiaozhi-server/core/utils/wakeup_word.py

@@ -1,4 +1,5 @@
 import os
+import re
 import yaml
 import time
 import hashlib
@@ -88,33 +89,30 @@ class WakeupWordsConfig:
         voice = hashlib.md5(voice.encode()).hexdigest()
         """获取唤醒词回复配置"""
         config = self._load_config()
-        default_response = {
-            "voice": "default",
-            "file_path": "config/assets/wakeup_words.wav",
-            "time": 0,
-            "text": "哈啰啊，我是小智啦，声音好听的台湾女孩一枚，超开心认识你耶，最近在忙啥，别忘了给我来点有趣的料哦，我超爱听八卦的啦",
-        }
 
         if not config or voice not in config:
-            return default_response
+            return None
 
         # 检查文件大小
         file_path = config[voice]["file_path"]
         if not os.path.exists(file_path) or os.stat(file_path).st_size < (15 * 1024):
-            return default_response
+            return None
 
         return config[voice]
 
     def update_wakeup_response(self, voice: str, file_path: str, text: str):
         """更新唤醒词回复配置"""
         try:
+            # 过滤表情符号
+            filtered_text = re.sub(r'[\U0001F600-\U0001F64F\U0001F900-\U0001F9FF]', '', text)
+            
             config = self._load_config()
             voice_hash = hashlib.md5(voice.encode()).hexdigest()
             config[voice_hash] = {
                 "voice": voice,
                 "file_path": file_path,
                 "time": time.time(),
-                "text": text,
+                "text": filtered_text,
             }
             self._save_config(config)
         except Exception as e: