Refactor directories across services: rename `audio_generator` to `generators`, `asr` to `recognizers`, and update all import paths for consistency and improved module organization.

src/VoiceDialogue/api/core/config.py

@@ -17,7 +17,7 @@ class TTSConfigInitializer:
         }
 
         try:
-            from services.audio.audio_generator import tts_config_registry
+            from services.audio.generators import tts_config_registry
             config_count = len(tts_config_registry.get_all_configs())
 
             result.update({

src/VoiceDialogue/api/core/lifespan.py

@@ -4,7 +4,7 @@ from contextlib import asynccontextmanager
 
 from fastapi import FastAPI
 
-from services.audio.audio_generator import tts_config_registry
+from services.audio.generators import tts_config_registry
 from utils import get_system_language
 from .config import TTSConfigInitializer
 from .service_factories import get_core_voice_service_definitions

src/VoiceDialogue/api/core/service_factories.py

@@ -3,9 +3,9 @@ from core.constants import (
     audio_frames_queue, user_voice_queue
 )
 from services.audio import EchoCancellingAudioCapture, TTSAudioGenerator, AudioStreamPlayer
-from services.audio.audio_generator import BaseTTSConfig, tts_config_registry
+from services.audio.generators import BaseTTSConfig, tts_config_registry
 from services.speech import SpeechStateMonitor, ASRWorker
-from services.text.text_generator import LLMResponseGenerator
+from services.text.generator import LLMResponseGenerator
 from .service_manager import ServiceDefinition
 
 

src/VoiceDialogue/api/dependencies/model_deps.py

@@ -37,7 +37,7 @@ def get_language_model(model_name: Optional[str] = None):
 def get_voice_model(speaker_name: str = "沈逸"):
     """获取语音模型依赖"""
     try:
-        from services.audio.audio_generator.voice_model import voice_model_registry
+        from services.audio.generators.voice_model import voice_model_registry
 
         speaker_mapping = {
             '罗翔': 0,

src/VoiceDialogue/api/routes/asr_routes.py

@@ -2,7 +2,7 @@ import logging
 
 from fastapi import APIRouter, HTTPException, Request, BackgroundTasks
 
-from services.speech.asr import asr_manager
+from services.speech.recognizers import asr_manager
 from ..core.service_factories import get_asr_worker_service_definition
 from ..schemas.asr_schemas import (
     SupportedLanguagesResponse, ASRInstanceRequest, ASRInstanceResponse

src/VoiceDialogue/api/routes/tts_routes.py

@@ -3,7 +3,7 @@ from typing import Optional
 
 from fastapi import APIRouter, HTTPException, BackgroundTasks, Request
 
-from services.audio.audio_generator import tts_config_registry
+from services.audio.generators import tts_config_registry
 from ..core.service_factories import get_tts_audio_generator_service_definition
 from ..schemas.tts_schemas import (
     TTSModelInfo, TTSModelListResponse, TTSModelLoadRequest,

src/VoiceDialogue/main.py

@@ -12,13 +12,13 @@ from core.constants import (
     text_input_queue,
     audio_output_queue
 )
-from services.audio.aec_audio_capture import EchoCancellingAudioCapture
-from services.audio.audio_answer import TTSAudioGenerator
-from services.audio.audio_generator.models import tts_config_registry
-from services.audio.audio_player import AudioStreamPlayer
-from services.speech.asr_service import ASRWorker
-from services.speech.speech_monitor import SpeechStateMonitor
-from services.text.text_generator import LLMResponseGenerator
+from services.audio.capture import EchoCancellingAudioCapture
+from services.audio.generator import TTSAudioGenerator
+from services.audio.generators.models import tts_config_registry
+from services.audio.player import AudioStreamPlayer
+from services.speech.monitor import SpeechStateMonitor
+from services.speech.recognizer import ASRWorker
+from services.text.generator import LLMResponseGenerator
 
 HERE = Path(__file__).parent
 language: typing.Literal['zh', 'en'] = 'en'

src/VoiceDialogue/services/audio/__init__.py

@@ -1,6 +1,6 @@
-from .aec_audio_capture import EchoCancellingAudioCapture
-from .audio_answer import TTSAudioGenerator
-from .audio_player import AudioStreamPlayer
+from .capture import EchoCancellingAudioCapture
+from .generator import TTSAudioGenerator
+from .player import AudioStreamPlayer
 
 __all__ = (
     "EchoCancellingAudioCapture",

src/VoiceDialogue/services/audio/{audio_answer.py → generator.py}

@@ -5,7 +5,7 @@ from queue import Empty
 from core.base import BaseThread
 from core.constants import dropped_audio_cache, user_still_speaking_event, voice_state_manager
 from models.voice_task import VoiceTask
-from .audio_generator import tts_manager, BaseTTSConfig
+from .generators import tts_manager, BaseTTSConfig
 
 
 class TTSAudioGenerator(BaseThread):

src/VoiceDialogue/services/audio/{audio_generator → generators}/configs/moyoyo.py

@@ -1,4 +1,4 @@
-from services.audio.audio_generator.models.moyoyo import MoYoYoTTSConfig
+from services.audio.generators.models.moyoyo import MoYoYoTTSConfig
 
 # 基础预训练模型文件映射
 BASE_PRETRAINED_FILES = {

src/VoiceDialogue/services/audio/{audio_generator → generators}/runtime/kokoro.py

@@ -3,9 +3,9 @@ from typing import Tuple, Optional
 import numpy as np
 from kokoro_onnx import Kokoro
 
-from services.audio.audio_generator.configs.kokoro import KokoroTTSConfig
-from services.audio.audio_generator.manager import tts_tables
-from services.audio.audio_generator.runtime.interface import TTSInterface
+from services.audio.generators.configs.kokoro import KokoroTTSConfig
+from services.audio.generators.manager import tts_tables
+from services.audio.generators.runtime.interface import TTSInterface
 
 
 @tts_tables.register("tts_classes", "kokoro")

src/VoiceDialogue/services/audio/{audio_generator → generators}/runtime/moyoyo.py

@@ -4,9 +4,9 @@ from typing import Tuple
 import numpy as np
 
 from config.paths import load_third_party
-from services.audio.audio_generator.manager import tts_tables
-from services.audio.audio_generator.models.moyoyo import MoYoYoTTSConfig
-from services.audio.audio_generator.runtime.interface import TTSInterface
+from services.audio.generators.manager import tts_tables
+from services.audio.generators.models.moyoyo import MoYoYoTTSConfig
+from services.audio.generators.runtime.interface import TTSInterface
 
 load_third_party()
 

src/VoiceDialogue/services/speech/__init__.py

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

src/VoiceDialogue/services/speech/{asr_service.py → recognizer.py}

@@ -8,7 +8,7 @@ from core.base import BaseThread
 from core.constants import user_still_speaking_event, voice_state_manager, dropped_audio_cache
 from models.voice_task import VoiceTask
 from utils.cache import LRUCacheDict
-from .asr import asr_manager
+from .recognizers import asr_manager
 
 
 class ASRWorker(BaseThread):

src/VoiceDialogue/services/speech/{asr → recognizers}/models/funasr.py

@@ -5,9 +5,9 @@ import numpy as np
 from funasr_onnx import SeacoParaformer, CT_Transformer
 
 from config import paths
-from services.speech.asr.manager import asr_tables
-from services.speech.asr.models.base import ASRInterface
-from services.speech.asr.utils import ensure_minimum_audio_duration
+from services.speech.recognizers.manager import asr_tables
+from services.speech.recognizers.models.base import ASRInterface
+from services.speech.recognizers.utils import ensure_minimum_audio_duration
 
 
 @asr_tables.register('asr_classes', 'funasr')

src/VoiceDialogue/services/speech/{asr → recognizers}/models/whisper.py

@@ -4,9 +4,9 @@ import numpy as np
 from pywhispercpp.model import Model
 
 from config import paths
-from services.speech.asr.manager import asr_tables
-from services.speech.asr.models.base import ASRInterface
-from services.speech.asr.utils import ensure_minimum_audio_duration
+from services.speech.recognizers.manager import asr_tables
+from services.speech.recognizers.models.base import ASRInterface
+from services.speech.recognizers.utils import ensure_minimum_audio_duration
 
 
 @asr_tables.register('asr_classes', 'whisper')

src/VoiceDialogue/services/text/{text_generator.py → generator.py}

@@ -9,7 +9,7 @@ from config import paths
 from core.base import BaseThread
 from core.constants import chat_history_cache
 from models.voice_task import VoiceTask
-from services.text.langchain_llm import preprocess_sentence_text, \
+from services.text.processor import preprocess_sentence_text, \
     create_langchain_chat_llamacpp_instance, create_langchain_pipeline, warmup_langchain_pipeline
 
 CHINESE_SYSTEM_PROMPT = ("你是善于模拟真实的思考过程的AI助手。"