522 lines
16 KiB
Python
522 lines
16 KiB
Python
"""
|
|
语音识别和合成插件
|
|
为EG引擎提供语音识别和文本转语音功能
|
|
"""
|
|
|
|
import sys
|
|
import os
|
|
import time
|
|
import threading
|
|
from typing import Dict, Any, Callable, List
|
|
|
|
# 添加项目根目录到Python路径
|
|
sys.path.append(os.path.dirname(os.path.abspath(__file__)))
|
|
|
|
from core.speech_manager import SpeechManager
|
|
from recognition.speech_recognizer import SpeechRecognizer
|
|
from synthesis.speech_synthesizer import SpeechSynthesizer
|
|
from utils.audio_utils import AudioUtils
|
|
|
|
class SpeechRecognitionSynthesisPlugin:
|
|
"""
|
|
语音识别和合成插件主类
|
|
提供完整的语音处理功能，包括语音识别和文本转语音合成
|
|
"""
|
|
|
|
def __init__(self, world):
|
|
"""
|
|
初始化语音识别和合成插件
|
|
|
|
Args:
|
|
world: EG引擎世界对象
|
|
"""
|
|
self.world = world
|
|
self.plugin_name = "speech_recognition_synthesis"
|
|
self.version = "1.0.0"
|
|
self.author = "EG Team"
|
|
|
|
# 插件状态
|
|
self.enabled = False
|
|
self.initialized = False
|
|
self.is_listening = False
|
|
|
|
# 核心组件
|
|
self.speech_manager = None
|
|
self.speech_recognizer = None
|
|
self.speech_synthesizer = None
|
|
self.audio_utils = None
|
|
|
|
# 配置参数
|
|
self.config = {
|
|
'sample_rate': 16000,
|
|
'channels': 1,
|
|
'buffer_size': 1024,
|
|
'language': 'zh-CN',
|
|
'voice': 'default',
|
|
'volume': 1.0,
|
|
'speed': 1.0,
|
|
'pitch': 1.0
|
|
}
|
|
|
|
# 回调函数
|
|
self.speech_recognized_callback = None
|
|
self.speech_synthesis_callback = None
|
|
self.error_callback = None
|
|
|
|
# 统计信息
|
|
self.stats = {
|
|
'recognized_count': 0,
|
|
'synthesized_count': 0,
|
|
'recognition_errors': 0,
|
|
'synthesis_errors': 0,
|
|
'processing_time': 0.0
|
|
}
|
|
|
|
# 线程管理
|
|
self.recognition_thread = None
|
|
self.synthesis_thread = None
|
|
self.running = False
|
|
|
|
print(f"✓ 语音识别和合成插件 v{self.version} 已创建")
|
|
|
|
def initialize(self) -> bool:
|
|
"""
|
|
初始化插件
|
|
|
|
Returns:
|
|
是否初始化成功
|
|
"""
|
|
try:
|
|
# 初始化核心组件
|
|
self.speech_manager = SpeechManager(self)
|
|
self.speech_recognizer = SpeechRecognizer(self)
|
|
self.speech_synthesizer = SpeechSynthesizer(self)
|
|
self.audio_utils = AudioUtils(self)
|
|
|
|
# 初始化各组件
|
|
if not self.speech_manager.initialize():
|
|
print("✗ 语音管理器初始化失败")
|
|
return False
|
|
|
|
if not self.speech_recognizer.initialize():
|
|
print("✗ 语音识别器初始化失败")
|
|
return False
|
|
|
|
if not self.speech_synthesizer.initialize():
|
|
print("✗ 语音合成器初始化失败")
|
|
return False
|
|
|
|
self.initialized = True
|
|
print("✓ 语音识别和合成插件初始化完成")
|
|
return True
|
|
|
|
except Exception as e:
|
|
print(f"✗ 插件初始化失败: {e}")
|
|
import traceback
|
|
traceback.print_exc()
|
|
return False
|
|
|
|
def enable(self) -> bool:
|
|
"""
|
|
启用插件
|
|
|
|
Returns:
|
|
是否启用成功
|
|
"""
|
|
try:
|
|
if not self.initialized:
|
|
print("✗ 插件未初始化")
|
|
return False
|
|
|
|
# 启用各组件
|
|
if not self.speech_manager.enable():
|
|
print("✗ 语音管理器启用失败")
|
|
return False
|
|
|
|
if not self.speech_recognizer.enable():
|
|
print("✗ 语音识别器启用失败")
|
|
return False
|
|
|
|
if not self.speech_synthesizer.enable():
|
|
print("✗ 语音合成器启用失败")
|
|
return False
|
|
|
|
self.enabled = True
|
|
self.running = True
|
|
|
|
# 启动识别线程
|
|
self.recognition_thread = threading.Thread(target=self._recognition_worker, daemon=True)
|
|
self.recognition_thread.start()
|
|
|
|
print("✓ 语音识别和合成插件已启用")
|
|
return True
|
|
|
|
except Exception as e:
|
|
print(f"✗ 插件启用失败: {e}")
|
|
import traceback
|
|
traceback.print_exc()
|
|
return False
|
|
|
|
def disable(self):
|
|
"""禁用插件"""
|
|
try:
|
|
self.running = False
|
|
self.enabled = False
|
|
|
|
# 禁用各组件
|
|
if self.speech_synthesizer:
|
|
self.speech_synthesizer.disable()
|
|
|
|
if self.speech_recognizer:
|
|
self.speech_recognizer.disable()
|
|
|
|
if self.speech_manager:
|
|
self.speech_manager.disable()
|
|
|
|
print("✓ 语音识别和合成插件已禁用")
|
|
|
|
except Exception as e:
|
|
print(f"✗ 插件禁用失败: {e}")
|
|
import traceback
|
|
traceback.print_exc()
|
|
|
|
def finalize(self):
|
|
"""清理插件资源"""
|
|
try:
|
|
self.disable()
|
|
|
|
# 清理各组件
|
|
if self.speech_synthesizer:
|
|
self.speech_synthesizer.finalize()
|
|
self.speech_synthesizer = None
|
|
|
|
if self.speech_recognizer:
|
|
self.speech_recognizer.finalize()
|
|
self.speech_recognizer = None
|
|
|
|
if self.speech_manager:
|
|
self.speech_manager.finalize()
|
|
self.speech_manager = None
|
|
|
|
if self.audio_utils:
|
|
self.audio_utils = None
|
|
|
|
self.initialized = False
|
|
print("✓ 语音识别和合成插件资源已清理")
|
|
|
|
except Exception as e:
|
|
print(f"✗ 插件资源清理失败: {e}")
|
|
import traceback
|
|
traceback.print_exc()
|
|
|
|
def update(self, dt: float):
|
|
"""
|
|
更新插件状态
|
|
|
|
Args:
|
|
dt: 时间增量
|
|
"""
|
|
try:
|
|
if not self.enabled:
|
|
return
|
|
|
|
# 更新各组件
|
|
if self.speech_manager:
|
|
self.speech_manager.update(dt)
|
|
|
|
if self.speech_recognizer:
|
|
self.speech_recognizer.update(dt)
|
|
|
|
if self.speech_synthesizer:
|
|
self.speech_synthesizer.update(dt)
|
|
|
|
except Exception as e:
|
|
print(f"✗ 插件更新失败: {e}")
|
|
if self.error_callback:
|
|
self.error_callback(f"插件更新失败: {e}")
|
|
|
|
def _recognition_worker(self):
|
|
"""语音识别工作线程"""
|
|
try:
|
|
while self.running and self.enabled:
|
|
if self.is_listening and self.speech_recognizer:
|
|
# 执行语音识别
|
|
self.speech_recognizer.process_audio()
|
|
time.sleep(0.01) # 10ms延迟
|
|
|
|
except Exception as e:
|
|
print(f"✗ 语音识别工作线程错误: {e}")
|
|
if self.error_callback:
|
|
self.error_callback(f"语音识别工作线程错误: {e}")
|
|
|
|
def start_listening(self) -> bool:
|
|
"""
|
|
开始语音监听
|
|
|
|
Returns:
|
|
是否开始成功
|
|
"""
|
|
try:
|
|
if not self.enabled:
|
|
print("✗ 插件未启用")
|
|
return False
|
|
|
|
if self.speech_recognizer:
|
|
result = self.speech_recognizer.start_listening()
|
|
if result:
|
|
self.is_listening = True
|
|
print("✓ 开始语音监听")
|
|
return result
|
|
return False
|
|
|
|
except Exception as e:
|
|
print(f"✗ 开始语音监听失败: {e}")
|
|
if self.error_callback:
|
|
self.error_callback(f"开始语音监听失败: {e}")
|
|
return False
|
|
|
|
def stop_listening(self):
|
|
"""停止语音监听"""
|
|
try:
|
|
if self.speech_recognizer:
|
|
self.speech_recognizer.stop_listening()
|
|
|
|
self.is_listening = False
|
|
print("✓ 停止语音监听")
|
|
|
|
except Exception as e:
|
|
print(f"✗ 停止语音监听失败: {e}")
|
|
if self.error_callback:
|
|
self.error_callback(f"停止语音监听失败: {e}")
|
|
|
|
def synthesize_speech(self, text: str, callback: Callable = None) -> bool:
|
|
"""
|
|
合成语音
|
|
|
|
Args:
|
|
text: 要合成的文本
|
|
callback: 合成完成回调函数
|
|
|
|
Returns:
|
|
是否开始合成
|
|
"""
|
|
try:
|
|
if not self.enabled:
|
|
print("✗ 插件未启用")
|
|
return False
|
|
|
|
if self.speech_synthesizer:
|
|
result = self.speech_synthesizer.synthesize(text, callback)
|
|
if result:
|
|
self.stats['synthesized_count'] += 1
|
|
print(f"✓ 开始合成语音: {text[:50]}...")
|
|
return result
|
|
return False
|
|
|
|
except Exception as e:
|
|
print(f"✗ 语音合成失败: {e}")
|
|
self.stats['synthesis_errors'] += 1
|
|
if self.error_callback:
|
|
self.error_callback(f"语音合成失败: {e}")
|
|
return False
|
|
|
|
def set_speech_recognized_callback(self, callback: Callable):
|
|
"""
|
|
设置语音识别回调函数
|
|
|
|
Args:
|
|
callback: 回调函数
|
|
"""
|
|
self.speech_recognized_callback = callback
|
|
|
|
def set_speech_synthesis_callback(self, callback: Callable):
|
|
"""
|
|
设置语音合成回调函数
|
|
|
|
Args:
|
|
callback: 回调函数
|
|
"""
|
|
self.speech_synthesis_callback = callback
|
|
|
|
def set_error_callback(self, callback: Callable):
|
|
"""
|
|
设置错误回调函数
|
|
|
|
Args:
|
|
callback: 回调函数
|
|
"""
|
|
self.error_callback = callback
|
|
|
|
def set_language(self, language: str):
|
|
"""
|
|
设置语言
|
|
|
|
Args:
|
|
language: 语言代码 (如 'zh-CN', 'en-US')
|
|
"""
|
|
self.config['language'] = language
|
|
if self.speech_recognizer:
|
|
self.speech_recognizer.set_language(language)
|
|
if self.speech_synthesizer:
|
|
self.speech_synthesizer.set_language(language)
|
|
print(f"✓ 语言设置为: {language}")
|
|
|
|
def set_voice(self, voice: str):
|
|
"""
|
|
设置语音
|
|
|
|
Args:
|
|
voice: 语音名称
|
|
"""
|
|
self.config['voice'] = voice
|
|
if self.speech_synthesizer:
|
|
self.speech_synthesizer.set_voice(voice)
|
|
print(f"✓ 语音设置为: {voice}")
|
|
|
|
def set_volume(self, volume: float):
|
|
"""
|
|
设置音量
|
|
|
|
Args:
|
|
volume: 音量 (0.0-1.0)
|
|
"""
|
|
self.config['volume'] = max(0.0, min(1.0, volume))
|
|
if self.speech_synthesizer:
|
|
self.speech_synthesizer.set_volume(volume)
|
|
print(f"✓ 音量设置为: {volume}")
|
|
|
|
def set_speed(self, speed: float):
|
|
"""
|
|
设置语速
|
|
|
|
Args:
|
|
speed: 语速 (0.5-2.0)
|
|
"""
|
|
self.config['speed'] = max(0.5, min(2.0, speed))
|
|
if self.speech_synthesizer:
|
|
self.speech_synthesizer.set_speed(speed)
|
|
print(f"✓ 语速设置为: {speed}")
|
|
|
|
def set_pitch(self, pitch: float):
|
|
"""
|
|
设置音调
|
|
|
|
Args:
|
|
pitch: 音调 (0.5-2.0)
|
|
"""
|
|
self.config['pitch'] = max(0.5, min(2.0, pitch))
|
|
if self.speech_synthesizer:
|
|
self.speech_synthesizer.set_pitch(pitch)
|
|
print(f"✓ 音调设置为: {pitch}")
|
|
|
|
def get_stats(self) -> Dict[str, Any]:
|
|
"""
|
|
获取统计信息
|
|
|
|
Returns:
|
|
统计信息字典
|
|
"""
|
|
# 合并各组件的统计信息
|
|
stats = self.stats.copy()
|
|
|
|
if self.speech_manager:
|
|
stats.update(self.speech_manager.get_stats())
|
|
|
|
if self.speech_recognizer:
|
|
stats.update(self.speech_recognizer.get_stats())
|
|
|
|
if self.speech_synthesizer:
|
|
stats.update(self.speech_synthesizer.get_stats())
|
|
|
|
return stats
|
|
|
|
def is_speaking(self) -> bool:
|
|
"""
|
|
检查是否正在播放语音
|
|
|
|
Returns:
|
|
是否正在播放语音
|
|
"""
|
|
if self.speech_synthesizer:
|
|
return self.speech_synthesizer.is_speaking()
|
|
return False
|
|
|
|
def stop_speaking(self):
|
|
"""停止播放语音"""
|
|
if self.speech_synthesizer:
|
|
self.speech_synthesizer.stop_speaking()
|
|
print("✓ 停止播放语音")
|
|
|
|
def get_available_voices(self) -> List[str]:
|
|
"""
|
|
获取可用语音列表
|
|
|
|
Returns:
|
|
语音名称列表
|
|
"""
|
|
if self.speech_synthesizer:
|
|
return self.speech_synthesizer.get_available_voices()
|
|
return []
|
|
|
|
def get_available_languages(self) -> List[str]:
|
|
"""
|
|
获取可用语言列表
|
|
|
|
Returns:
|
|
语言代码列表
|
|
"""
|
|
if self.speech_recognizer and self.speech_synthesizer:
|
|
recognizer_langs = self.speech_recognizer.get_available_languages()
|
|
synthesizer_langs = self.speech_synthesizer.get_available_languages()
|
|
# 返回两个列表的交集
|
|
return list(set(recognizer_langs) & set(synthesizer_langs))
|
|
elif self.speech_recognizer:
|
|
return self.speech_recognizer.get_available_languages()
|
|
elif self.speech_synthesizer:
|
|
return self.speech_synthesizer.get_available_languages()
|
|
return []
|
|
|
|
def save_audio_file(self, filename: str, audio_data: bytes) -> bool:
|
|
"""
|
|
保存音频文件
|
|
|
|
Args:
|
|
filename: 文件名
|
|
audio_data: 音频数据
|
|
|
|
Returns:
|
|
是否保存成功
|
|
"""
|
|
if self.audio_utils:
|
|
return self.audio_utils.save_audio_file(filename, audio_data)
|
|
return False
|
|
|
|
def load_audio_file(self, filename: str) -> bytes:
|
|
"""
|
|
加载音频文件
|
|
|
|
Args:
|
|
filename: 文件名
|
|
|
|
Returns:
|
|
音频数据
|
|
"""
|
|
if self.audio_utils:
|
|
return self.audio_utils.load_audio_file(filename)
|
|
return b''
|
|
|
|
# 插件注册信息
|
|
def get_plugin_info():
|
|
"""获取插件信息"""
|
|
return {
|
|
'name': 'speech_recognition_synthesis',
|
|
'version': '1.0.0',
|
|
'author': 'EG Team',
|
|
'description': '语音识别和合成插件，提供语音识别和文本转语音功能',
|
|
'dependencies': []
|
|
}
|
|
|
|
# 创建插件实例的工厂函数
|
|
def create_plugin(world):
|
|
"""创建插件实例"""
|
|
return SpeechRecognitionSynthesisPlugin(world) |