smartpi 1.1.2__py3-none-any.whl → 1.1.4__py3-none-any.whl

smartpi/ai_vad.py

@@ -1,83 +0,0 @@
-# -*- coding: utf-8 -*-
-import sys
-import os
-import time
-import numpy as np
-
-class VADProviderBase:
-    """VAD provider base class"""
-    @staticmethod
-    def create_vad_instance(
-        vad_mini_volume=1000, 
-        vad_mini_length=500,
-        vad_pause_timeout=3000, 
-        vad_silence_timeout=10000,
-        threshold=0.02
-    ):
-        """Factory method to create VAD instance - always returns EnergyVADProvider"""
-        # 忽略vad_mini_volume, vad_mini_length, vad_silence_timeout参数
-        # 将vad_pause_timeout作为min_silence_duration_ms传递
-        return EnergyVADProvider(
-            threshold=threshold, 
-            min_silence_duration_ms=vad_pause_timeout
-        )
-
-class EnergyVADProvider(VADProviderBase):
-    """Audio energy-based VAD provider"""
-    def __init__(self, threshold=0.02, min_silence_duration_ms=300):
-        print(f"Initializing EnergyVAD: threshold={threshold}, min_silence_duration_ms={min_silence_duration_ms}")
-        # 初始化VAD配置，使用默认值作为备选
-        self.vad_threshold = float(threshold) if threshold else 0.02
-        self.silence_threshold_ms = int(min_silence_duration_ms) if min_silence_duration_ms else 300
-        print("EnergyVAD initialized successfully")
-    
-    def is_vad(self, conn, audio_data):
-        """使用音频能量检测语音活动"""
-        try:
-            # 确保conn对象有必要的属性
-            if not hasattr(conn, 'client_audio_buffer'):
-                conn.client_audio_buffer = []
-            if not hasattr(conn, 'client_have_voice'):
-                conn.client_have_voice = False
-            if not hasattr(conn, 'client_have_voice_last_time'):
-                conn.client_have_voice_last_time = 0
-            if not hasattr(conn, 'client_voice_stop'):
-                conn.client_voice_stop = False
-            
-            # 将新数据加入缓冲区
-            conn.client_audio_buffer.extend(audio_data)
-            
-            # 处理缓冲区中的数据
-            client_have_voice = False
-            # 对于能量检测，我们可以使用较小的块大小，例如256采样点
-            while len(conn.client_audio_buffer) >= 256 * 2:
-                # 提取前256个采样点（512字节）
-                chunk = bytes(conn.client_audio_buffer[:256 * 2])
-                conn.client_audio_buffer = conn.client_audio_buffer[256 * 2:]
-                
-                # 计算能量
-                audio_int16 = np.frombuffer(chunk, dtype=np.int16)
-                energy = np.abs(audio_int16).mean()
-                
-                client_have_voice = energy > self.vad_threshold
-                
-                # 更新VAD状态
-                if client_have_voice:
-                    conn.client_have_voice = True
-                    conn.client_have_voice_last_time = time.time() * 1000
-                    conn.client_voice_stop = False
-                elif conn.client_have_voice:
-                    # 检查静默时间
-                    stop_duration = time.time() * 1000 - conn.client_have_voice_last_time
-                    if stop_duration >= self.silence_threshold_ms:
-                        conn.client_have_voice = False
-                        conn.client_voice_stop = True
-            
-            return client_have_voice
-        except Exception as e:
-            print(f"能量检测出错: {e}")
-            return False
-    
-    def reset_states(self):
-        """重置VAD状态"""
-        pass  # 状态现在存储在conn对象中，不需要在这里重置

smartpi/audio.py

@@ -1,125 +0,0 @@
-#!/usr/bin/env python3
-import os
-import sys
-import subprocess
-import numpy as np
-import math
-
-# 全局导入，避免每次调用都重复导入
-import warnings
-warnings.filterwarnings('ignore')
-
-# 在导入sounddevice前设置环境变量抑制警告
-os.environ['PYTHONWARNINGS'] = 'ignore'
-
-# 导入sounddevice
-import sounddevice as sd
-
-def record(seconds, filename="recording.wav"):
-    """
-    录音函数
-    示例: record(10, "voice.wav")
-    """
-    cmd = f"sudo record.sh -d {seconds} {filename}"
-    print(f"录音{seconds}秒 -> {filename}")
-    result = subprocess.run(cmd, shell=True, capture_output=True, text=True)
-    if result.returncode == 0:
-        print(f"录音成功: {filename}")
-        return True
-    else:
-        print(f"录音失败: {result.stderr}")
-        return False
-
-def get_db(sample_time=0.1):
-    """
-    检测麦克风分贝值
-    返回: 0-100的整数分贝值
-    """
-    try:
-        # 获取可用设备
-        devices = sd.query_devices()
-        
-        # 选择第一个可用的输入设备
-        input_device = None
-        for i, dev in enumerate(devices):
-            if dev['max_input_channels'] > 0:
-                input_device = i
-                break
-        
-        if input_device is None:
-            return 0
-        
-        # 录音参数
-        samplerate = 44100
-        channels = 1
-        duration = sample_time
-        
-        # 录音
-        recording = sd.rec(
-            int(duration * samplerate),
-            samplerate=samplerate,
-            channels=channels,
-            device=input_device,
-            blocking=True,
-            dtype='int16'
-        )
-        
-        # 转换为numpy数组
-        audio_data = recording.flatten()
-        
-        # 过滤静音
-        audio_data = audio_data[np.abs(audio_data) > 50]
-        
-        if len(audio_data) == 0:
-            return 0
-        
-        # 计算RMS
-        rms = np.sqrt(np.mean(np.square(audio_data.astype(np.float32))))
-        
-        # 计算分贝值
-        if rms > 0:
-            db = 20 * math.log10(rms / 32767.0)
-            # 转换为0-100范围并取整
-            db_value = max(0, min(100, (db + 60) * (100/60)))
-            return int(db_value)
-        
-        return 0
-        
-    except Exception:
-        # 发生任何异常返回0
-        return 0
-
-def play(audio_file):
-    """
-    播放音频文件
-    示例: play("music.wav")
-    """
-    if not os.path.exists(audio_file):
-        print(f"文件不存在: {audio_file}")
-        return False
-    
-    try:
-        subprocess.run(f"aplay {audio_file}", shell=True, check=True)
-        return True
-    except subprocess.CalledProcessError as e:
-        print(f"播放失败: {str(e)}")
-        return False
-    except Exception as e:
-        print(f"播放错误: {str(e)}")
-        return False
-
-def volume(level=None):
-    """
-    调节音量百分比 (0-100)
-    示例: 
-      volume(80)    # 设置音量到80%
-      current = volume()  # 获取当前音量
-    """
-    
-    try:      
-        level = max(0, min(100, level))
-        subprocess.run(['amixer', '-q', 'sset', 'Master', f'{level}%'], stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL, check=True)
-        return level
-    except Exception as e:
-        print(f"音量操作错误: {str(e)}")
-        return 0 if level is None else level