1. intel Edison和arduino 扩展板

2. 1个intel Edison 电源

3. 1个麦克风 1个USB声卡（带音频输入输出）

4. 带迷你杰克头的电源控制的扬声器

5. 螺母和螺栓

6. 喷胶或胶布

根据附件的图纸，准备好外壳的各个部分，打印出装饰的图片，方便最后组装

1. 把A7和A8装到A1上

2. 用胶布或双面胶把彩色外观贴到A1，A2，A3上

3. 把按钮放到A1上设计好的孔里, 注意按钮的顺序要与LEDs匹配，把LED放进A1上设计好的孔里，注意LED的顺序要与按钮匹配

4. 在A4上安装Arduino扩展板

5. 连接声音元件、声卡、麦克风和扬声器、把电路装到Arduino 扩展板上

6. 把A5和A6连到A1上

7. 从上面的孔装入麦克风，从下面的孔装电线

8. 在A7和A8上插入A2和A3，用橡皮筋固定它们

9. 用A4完成组装，小心电线。在盖上最后的盖子前，最好先测试下电路是否能用

如果你有最新的intel Edison系统映像，那很可能你不需要装任何驱动，声卡就能用。插好USB声卡，打开开关。在打开开关前，请确保连好电源。否则，Intel Edison 无法加电

先用通用的 alsamixer检查下声卡能否正常工作和控制音量

在控制台输入“alsamixer”打开界面。按下F6，选择声卡。在这里，我是选的C-Media USB头戴式耳机设置。把输出音量调低，这样不会太吵。通过调节这个，找到合适的音量。

安装python库

升级设置工具，它是用来安装程序包的Python工具。确保你能连上网络，在命令控制台输入如下语句：

wget https://bitbucket.org/pypa/setuptools/raw/bootstrap/ez_setup.py -O - | python

下载pyaudio，这是我们要用来处理声音输入/输出的Python库。在命令控制台输入如下语句：

wget people.csail.mit.edu/hubert/pyaudio/packages/pyaudio-0.2.8.tar.gz -o ./pyaudio.tar.gz

在命令控制台输入如下语句解码：

tar -x pyaudio.tar.gz

安装pyaudio 根据导航进入pyaudio的文件夹，在命令控制台输入如下语句：

python setup.py install

安装portaudio，这是让音频设备正常工作的Linux工具。在命令控制台输入如下语句：

opkg install libportaudio2

安装numpy，这是处理数据的Python库，它会在音频处理的实际进程上用到。在命令控制台输入如下语句：

opkg install python-numpy

intel Edison的移动脚本

从附件下载python脚本，用sip把它全部复制到intel Edison

在命令控制台输入如下语句：

scp Mimic.py root@your-edison-ip:~/

在命令控制台开始编写小怪物米米卡的程序：

python ~/Mimic.py

源代码如下：

"""
Mimic:
Let Edison mimic your voice in a funny way

"""

import pyaudio
import audioop
import numpy as np
import struct
import wave
import time
import mraa

led_pins=[13,12,11,10]
btn_pins=[9,8,7,6]
levels=[-7,-3,3,7]
leds=[]
btns=[]

CHUNK = 1024
WIDTH = 2
CHANNELS = 1 
RATE = 44100
RECORD_SECONDS = 5
level=-1
swidth=2

#Volume needed to activate the programme
activate_volume=2000

def dimAllLeds():
    for i in range(4):
        leds[i].write(0)

def speedx(snd_array, factor):
    """ Speeds up / slows down a sound, by some factor. """
    indices = np.round( np.arange(0, len(snd_array), factor) )
    indices = indices[indices < len(snd_array)].astype(int)
    return snd_array[ indices ]

def stretch(snd_array, factor, window_size, h):
    """ Stretches/shortens a sound, by some factor. """
    phase  = np.zeros(window_size)
    hanning_window = np.hanning(window_size)
    result = np.zeros( len(snd_array) /factor + window_size)

    for i in np.arange(0, len(snd_array)-(window_size+h), h*factor):

        # two potentially overlapping subarrays
        a1 = snd_array[i: i + window_size]
        a2 = snd_array[i + h: i + window_size + h]

        # the spectra of these arrays
        s1 =  np.fft.fft(hanning_window * a1)
        s2 =  np.fft.fft(hanning_window * a2)

        #  rephase all frequencies
        phase = (phase + np.angle(s2/s1)) % 2*np.pi

        a2_rephased = np.fft.ifft(np.abs(s2)*np.exp(1j*phase))
        i2 = int(i/factor)
        result[i2 : i2 + window_size] += hanning_window*a2_rephased

    result = ((2**(16-4)) * result/result.max()) # normalize (16bit)

    return result.astype('int16')

def pitchshift(snd_array, n, window_size=2**13, h=2**11):
    """ Changes the pitch of a sound by ``n`` semitones. """
    factor = 2**(1.0 * n / 12.0)
    stretched = stretch(snd_array, 1.0/factor, window_size, h)
    return speedx(stretched[window_size:], factor)


if __name__=="__main__":
    #Initialize the buttons and leds
    for i in range(0,4):
        led=mraa.Gpio(led_pins[i])
        led.dir(mraa.DIR_OUT)
        btn=mraa.Gpio(btn_pins[i])
        btn.dir(mraa.DIR_IN)
        btn.mode(mraa.MODE_PULLUP)
        leds.append(led)
        btns.append(btn)

    #Initialize audio stream
    p = pyaudio.PyAudio()
    stream = p.open(format=p.get_format_from_width(WIDTH),
                    channels=CHANNELS,
                    rate=RATE,
                    input=True,
                    output=True,
                    frames_per_buffer=CHUNK)



    #The main loop
    while True:
        frames=""
        stream.start_stream()
        
        #Work to do before mimic starts
        while True:
            #Check if buttons are pressed
            for i in range(4):
                if btns[i].read()==0:
                    dimAllLeds()
                    leds[i].write(1)
                    level=levels[i]
            #Check if someone is speaking 
            data = stream.read(CHUNK)
            rms=audioop.rms(data,2)
            if rms>activate_volume:
                frames=frames+data
                break
                
        #Start recording
        timer=0
        print("* recording")
        while True:
            try:
                data = stream.read(CHUNK)
                frames=frames+data
                rms=audioop.rms(data,2)
                if rms<activate_volume:
                    if time.time()-timer>1:
                        break
                else:
                    timer=time.time()
            except:
                pass
                
        #Do the pitch shifting magic
        frames = struct.unpack("%dh"%(len(frames)/swidth), frames)
        frames = np.array(frames)
        frames=pitchshift(frames, level)
        frames=struct.pack("%dh"%(len(frames)),*list(frames))
        
        #Play the modified sound
        data = frames[0:CHUNK] 
        i=0
        while data != '':
            stream.write(data)
            i=i+1
            data = frames[i*CHUNK:(i+1)*CHUNK] 

        print("* done")
        stream.stop_stream()
        
    stream.close()
    p.terminate()

对准麦克风清楚大声的说话，小怪物米米卡就会用搞笑的声音重复你的话了~~~~