Add system utilities and initialize core modules: implement `get_system_language` and `get_system_info`, update API startup with system defaults, and integrate ASR, LLM, and speech modules for enhanced functionality.

src/VoiceDialogue/api/app.py

@@ -5,6 +5,13 @@ from typing import Dict, Any
 from fastapi import FastAPI, HTTPException, APIRouter
 from fastapi.middleware.cors import CORSMiddleware
 
+from services.audio.audio_player import AudioStreamPlayer
+from services.core.constants import (
+    transcribed_text_queue, text_input_queue, audio_output_queue, audio_frames_queue, user_voice_queue
+)
+from services.speech import SpeechStateMonitor, ASRWorker
+from services.text.text_generator import LLMResponseGenerator
+from utils import get_system_language
 from .middleware.logging import LoggingMiddleware
 from .middleware.rate_limit import RateLimitMiddleware
 from .routes import tts_routes, asr_routes
@@ -26,6 +33,8 @@ async def lifespan(app: FastAPI):
     # 启动时的初始化
     logger.info("正在启动VoiceDialogue API服务...")
 
+    system_default_language = get_system_language()
+
     # 初始化TTS配置注册表
     try:
         from services.audio.audio_generator import tts_config_registry
@@ -35,6 +44,30 @@ async def lifespan(app: FastAPI):
         logger.error(f"TTS配置加载失败: {e}")
         app_state["tts_configs_loaded"] = False
 
+    speech_monitor = SpeechStateMonitor(
+        audio_frame_queue=audio_frames_queue,
+        user_voice_queue=user_voice_queue,
+    )
+    speech_monitor.start()
+    app_state['speech_monitor'] = speech_monitor
+
+    asr_worker = ASRWorker(
+        user_voice_queue=user_voice_queue, transcribed_text_queue=transcribed_text_queue,
+        language=system_default_language
+    )
+    asr_worker.start()
+    app_state['asr_worker'] = asr_worker
+
+    llm_generator = LLMResponseGenerator(
+        user_question_queue=transcribed_text_queue, generated_answer_queue=text_input_queue
+    )
+    llm_generator.start()
+    app_state['llm_generator'] = llm_generator
+
+    audio_player = AudioStreamPlayer(audio_playing_queue=audio_output_queue)
+    audio_player.start()
+    app_state['audio_player'] = audio_player
+
     app_state["system_running"] = True
     logger.info("VoiceDialogue API服务启动完成")
     yield

src/VoiceDialogue/services/speech/__init__.py

@@ -0,0 +1,4 @@
+from .asr_service import ASRWorker
+from .speech_monitor import SpeechStateMonitor
+
+__all__ = ['ASRWorker', 'SpeechStateMonitor']

src/VoiceDialogue/utils/__init__.py

@@ -1,26 +1,26 @@
+from .cache import LRUCacheDict
 from .download_utils import (
     download_model_from_huggingface, download_file_from_huggingface, check_file_exists_on_huggingface,
     download_lora_from_huggingface, download_civitai_file
 )
 from .strings import remove_emojis
-from .cache import LRUCacheDict
-
+from .system import get_system_language, get_system_info
 
 # 导入HParams类，解决moyoyo_tts的序列化问题
 try:
     import sys
     from pathlib import Path
-    
+
     # 添加third_party路径
     current_dir = Path(__file__).parent
     project_root = current_dir.parent.parent.parent
     third_party_path = project_root / "third_party"
-    
+
     if str(third_party_path) not in sys.path:
         sys.path.insert(0, str(third_party_path))
-    
+
     from moyoyo_tts.utils import HParams
-    
+
 except ImportError:
     # 如果导入失败，创建一个简单的HParams类
     class HParams:
@@ -54,7 +54,6 @@ except ImportError:
         def __repr__(self):
             return self.__dict__.__repr__()
 
-
 __all__ = (
     'remove_emojis',
     'download_model_from_huggingface',
@@ -62,4 +61,7 @@ __all__ = (
     'check_file_exists_on_huggingface',
     'download_lora_from_huggingface',
     'download_civitai_file',
+    'LRUCacheDict',
+    'get_system_language',
+    'get_system_info',
 )

src/VoiceDialogue/utils/system.py

@@ -0,0 +1,68 @@
+"""系统相关工具函数"""
+
+import locale
+import os
+import platform
+from typing import Literal
+
+__all__ = ('get_system_language', 'get_system_info')
+
+
+def get_system_language() -> Literal['zh', 'en']:
+    """
+    获取系统当前语言设置
+    
+    Returns:
+        str: 返回 'zh' 或 'en'，默认为 'zh'
+    """
+    try:
+        # 首先尝试从环境变量获取
+        lang = os.environ.get('LANG', '')
+        if not lang:
+            lang = os.environ.get('LANGUAGE', '')
+        if not lang:
+            lang = os.environ.get('LC_ALL', '')
+        if not lang:
+            lang = os.environ.get('LC_MESSAGES', '')
+
+        # 如果环境变量为空，尝试使用locale模块
+        if not lang:
+            try:
+                lang, _ = locale.getdefaultlocale()
+                if not lang:
+                    lang = locale.getlocale()[0]
+            except (ValueError, AttributeError):
+                lang = None
+
+        # 解析语言代码
+        if lang:
+            lang = lang.lower()
+            if 'zh' in lang or 'chinese' in lang:
+                return 'zh'
+            elif 'en' in lang or 'english' in lang:
+                return 'en'
+
+    except Exception:
+        # 如果所有方法都失败，返回默认值
+        pass
+
+    # 默认返回中文
+    return 'zh'
+
+
+def get_system_info() -> dict:
+    """
+    获取系统信息
+    
+    Returns:
+        dict: 包含系统信息的字典
+    """
+    return {
+        'platform': platform.system(),
+        'platform_version': platform.version(),
+        'architecture': platform.architecture()[0],
+        'machine': platform.machine(),
+        'processor': platform.processor(),
+        'language': get_system_language(),
+        'python_version': platform.python_version(),
+    }