def main():
global mic, quit_event
bing = BingVoice(BING_KEY)
awake = False
pa = pyaudio.PyAudio()
mic = Microphone(pa)
player = Player(pa)
while not quit_event.is_set():
if not awake:
if mic.recognize(keyword='hey respeaker'):
awake = True
player.play(hi)
continue
else:
break
command = mic.recognize(max_phrase_ms=6000, max_wait_ms=6000)
if command:
print('recognized: ' + command)
if command.find('play music') > 0:
pass
awake = False
mic.close()
def __init__(self):
self.sig_hndlr = signal.signal(signal.SIGINT, self.exit_gracefully)
#set up GPIO
self.driveLeft = PWMOutputDevice(PWM_DRIVE_LEFT, True, 0, 1000)
self.driveRight = PWMOutputDevice(PWM_DRIVE_RIGHT, True, 0, 1000)
self.forwardLeft = DigitalOutputDevice(FORWARD_LEFT_PIN)
self.reverseLeft = DigitalOutputDevice(REVERSE_LEFT_PIN)
self.forwardRight = DigitalOutputDevice(FORWARD_RIGHT_PIN)
self.reverseRight = DigitalOutputDevice(REVERSE_RIGHT_PIN)
# Set up sensors
self.camera = Camera()
self.microphone = Microphone()
self.state = Idle()
class AudioVideo:
"""
Audio/Video class. Combines motion detection with recording
of audio and video, then stitches together .h264 video with
.wav audio to make .mp4 video
"""
def __init__(self):
self.mic = Microphone()
self.cam = Camera()
def startRecording(self):
self.start_time = time.time()
self.mic.start()
self.cam.start()
def stopRecording(self):
self.mic.stop()
self.cam.stop()
print("Recording stops.")
print("--- %s seconds ---" % (time.time() - self.start_time))
def main():
global mic, quit_event
bing = BingVoice(BING_KEY)
awake = False
pa = pyaudio.PyAudio()
mic = Microphone(pa)
player = Player(pa)
while not quit_event.is_set():
if not awake:
if mic.recognize(keyword='hey respeaker'):
awake = True
player.play(hi)
continue
else:
break
data = b''.join(mic.listen())
if data:
# recognize speech using Microsoft Bing Voice Recognition
try:
text = bing.recognize(data, language='en-US')
print('Bing:' + text.encode('utf-8'))
tts_data = bing.synthesize('you said ' + text)
player.play_raw(tts_data)
if text.find('start recording') >= 0:
mic.record('record.wav')
elif text.find('stop recording') >= 0:
mic.interrupt(stop_recording=True)
elif text.find('play recording audio') >= 0:
player.play('record.wav')
except UnknownValueError:
print("Microsoft Bing Voice Recognition could not understand audio")
except RequestError as e:
print("Could not request results from Microsoft Bing Voice Recognition service; {0}".format(e))
else:
print('no data')
awake = False
mic.close()
class Client():
def __init__(self, master):
self.microphone = Microphone()
self.audioToNet = AudioToNet()
self.trafficPrinter = TrafficPrinter()
self.master = master
def call(self, host, port):
self.microphone.init()
s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
self.master.serveOn(s)
self.sendTo(s, host, port)
self.microphone.destroy()
def sendTo(self, socket, host, port):
print "Sending to {0}, port {1}".format(host, port)
while True:
audio = self.microphone.nextAudioChunk()
chunk = self.audioToNet.convert(audio)
self.sendChunk(chunk, socket, host, port)
def sendChunk(self, chunk, socket, host, port):
socket.sendto(chunk, (host, port))
self.trafficPrinter.addTraffic(len(chunk))
class Client():
def __init__(self, master):
self.microphone = Microphone()
self.audioToNet = AudioToNet()
self.trafficPrinter = TrafficPrinter()
self.master = master
def call(self, host, port):
self.microphone.init()
s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
self.master.serveOn(s)
self.sendTo(s, host, port)
self.microphone.destroy()
def sendTo(self, socket, host, port):
print "Sending to {0}, port {1}".format(host, port)
while True:
audio = self.microphone.nextAudioChunk()
chunk = self.audioToNet.convert(audio)
self.sendChunk(chunk, socket, host, port)
def sendChunk(self, chunk, socket, host, port):
socket.sendto(chunk, (host, port))
self.trafficPrinter.addTraffic(len(chunk))
from worker import Worker
import time
script_dir = os.path.dirname(os.path.realpath(__file__))
hi = os.path.join(script_dir, 'audio/hi.wav')
bing = BingBase(BING_KEY)
recognizer = BingVoiceRecognizer(bing)
tts = BingTTS(bing)
mission_completed = False
awake = False
pa = pyaudio.PyAudio()
mic = Microphone(pa)
player = Player(pa)
mic.player = player
worker = Worker()
worker.set_tts(tts)
worker.set_player(player)
def handle_int(sig, frame):
global mission_completed
print "Terminating..."
mission_completed = True
mic.close()
player.close()
from worker import Worker
import time
script_dir = os.path.dirname(os.path.realpath(__file__))
hi = os.path.join(script_dir, 'audio/hi.wav')
bing = BingBase(BING_KEY)
recognizer = BingVoiceRecognizer(bing)
tts = BingTTS(bing)
mission_completed = False
awake = False
pa = pyaudio.PyAudio()
mic = Microphone(pa)
player = Player(pa)
worker = Worker()
worker.set_tts(tts)
worker.set_player(player)
def handle_int(sig, frame):
global mission_completed
print "Terminating..."
mission_completed = True
mic.close()
player.close()
worker.stop()
pa.terminate()
sys.exit(-1)
script_dir = os.path.dirname(os.path.realpath(__file__))
hi = os.path.join(script_dir, 'audio/hi.wav')
spi = SPI()
spi.write('offline\n')
bing = BingVoice(BING_KEY)
mission_completed = False
awake = False
pa = pyaudio.PyAudio()
mic = Microphone(pa)
player = Player(pa)
def check_internet(host="8.8.8.8", port=53, timeout=6):
"""
Host: 8.8.8.8 (google-public-dns-a.google.com)
OpenPort: 53/tcp
Service: domain (DNS/TCP)
"""
try:
socket.setdefaulttimeout(timeout)
socket.socket(socket.AF_INET, socket.SOCK_STREAM).connect((host, port))
return True
except Exception as ex:
print ex.message
def __init__(self, master): self.microphone = Microphone() self.audioToNet = AudioToNet() self.trafficPrinter = TrafficPrinter() self.master = master
def __init__(self):
self.mic = Microphone()
self.cam = Camera()
class Robot(State):
def __init__(self):
self.sig_hndlr = signal.signal(signal.SIGINT, self.exit_gracefully)
#set up GPIO
self.driveLeft = PWMOutputDevice(PWM_DRIVE_LEFT, True, 0, 1000)
self.driveRight = PWMOutputDevice(PWM_DRIVE_RIGHT, True, 0, 1000)
self.forwardLeft = DigitalOutputDevice(FORWARD_LEFT_PIN)
self.reverseLeft = DigitalOutputDevice(REVERSE_LEFT_PIN)
self.forwardRight = DigitalOutputDevice(FORWARD_RIGHT_PIN)
self.reverseRight = DigitalOutputDevice(REVERSE_RIGHT_PIN)
# Set up sensors
self.camera = Camera()
self.microphone = Microphone()
self.state = Idle()
def on_event(self, event):
self.state = self.state.on_event(event)
def allStop(self):
self.forwardLeft.value = False
self.reverseLeft.value = False
self.forwardRight.value = False
self.reverseRight.value = False
self.driveLeft.value = 0
self.driveRight.value = 0
def goForward(self):
self.forwardLeft.value = True
self.reverseLeft.value = False
self.forwardRight.value = True
self.reverseRight.value = False
self.driveLeft.value = 1.0
self.driveRight.value = 1.0
def goBackward(self):
self.forwardLeft.value = False
self.everseLeft.value = True
self.orwardRight.value = False
self.reverseRight.value = True
self.driveLeft.value = 1.0
self.driveRight.value = 1.0
def rotateRight(self):
self.forwardLeft.value = True
self.reverseLeft.value = False
self.forwardRight.value = False
self.reverseRight.value = True
self.driveLeft.value = 0.8
self.driveRight.value = 0.8
def rotateLeft(self):
self.forwardLeft.value = False
self.reverseLeft.value = True
self.forwardRight.value = True
self.reverseRight.value = False
self.driveLeft.value = 0.8
self.driveRight.value = 0.8
def bankRight(self):
self.forwardLeft.value = True
self.reverseLeft.value = False
self.forwardRight.value = True
self.reverseRight.value = False
self.driveLeft.value = 0.8
self.driveRight.value = 0.2
def bankLeft(self):
self.forwardLeft.value = True
self.reverseLeft.value = False
self.forwardRight.value = True
self.reverseRight.value = False
self.driveLeft.value = 0.2
self.driveRight.value = 0.8
def detectHuman(self):
hog = cv2.HOGDescriptor()
hog.setSVMDetector(cv2.HOGDescriptor_getDefaultPeopleDetector())
im = self.camera.read()
im = imutils.resize(im, width=min(400, im.shape[1]))
orig = im.copy()
#detect peope in the image
(rects, weights) = hog.detectMultiScale(im, winStride=(4, 4), padding=(8, 8), scale=1.05)
# draw the original bounding boxes
for (x, y, w, h) in rects:
cv2.rectangle(orig, (x, y), (x + w, y + h), (0, 0, 255), 2)
# apply non-maxima suppression to the bounding boxes using a fiarly large overlap
# threshold to try to maintain overlapping
rects = np.array([[x, y, x + w, y + h] for (x, y, w, h) in rects])
pick = non_max_suppression(rects, probs=None, overlapThresh=0.65)
# draw final bounding boxes
bounded_box = None
for (xA, yA, xB, yB) in pick:
# here we assume one person is in the frame
bounded_box = (xA, yA, xB, yB)
cv2.rectangle(im, (xA, yA), (xB, yB), (0, 255,0), 2)
# show images
#cv2.imshow("Before", orig)
cv2.imshow("After", im)
print bounded_box
return bounded_box
def followHuman(self):
# Variables to Change
binsize = 17
turningTime = 0.2
midrange_L = 7
midrange_R = 11
bounded_box = self.detectHuman()
if bounded_box == None:
return
mid_pt = (bounded_box[0] + bounded_box[2]) / 2
if (mid_pt > (400*midrange_L) / binsize and mid_pt < (400*midrange_R) / binsize):
self.goForward()
time.sleep(1.0)
self.allStop()
print "human in view"
elif (mid_pt < (400*midrange_L) / binsize):
print "left side"
self.rotateLeft()
time.sleep(turningTime)
self.allStop()
else:
print "right side"
self.rotateRight()
time.sleep(turningTime)
self.allStop()
def startupSensors(self):
self.camera.startStream()
self.microphone.startRecording()
time.sleep(1)
def quit(self):
self.camera.close()
self.microphone.close()
self.allStop()
def runStateMachine(self):
Thread(target=self.stateMachine, args=()).start()
def runRobot(self):
while True:
print "Recording"
if (self.microphone.startRecording()):
print "Done Recording"
self.on_event(self.microphone.read())
print self.state.__str__()
print self.state.__str__() == 'Idle'
if (self.state.__str__() == 'Idle'):
self.allStop()
elif (self.state.__str__() == 'FollowHumanIdle'):
#self.camera.startStream()
self.allStop()
elif (self.state.__str__() == 'FollowHuman'):
self.followHuman()
elif (self.state.__str__() == 'VoiceControlIdle'):
print "VCIdle"
self.allStop()
elif (self.state.__str__() == 'Forward'):
print "for"
self.goForward()
elif (self.state.__str__() == 'RotateLeft'):
print "RLeft"
self.rotateLeft()
elif (self.state.__str__() == 'RotateRight'):
self.rotateRight()
elif (self.state.__str__() == 'QuitApp'):
self.quit()
return
time.sleep(2.0)
def followSpeech(self):
while True:
if (self.microphone.startRecording()):
print "Done Recording"
print self.microphone.read()
self.state = self.microphone.read()
if (self.state.__str__() == "go"):
self.goForward()
elif(self.state.__str__() == "left"):
self.rotateLeft()
elif(self.state.__str__() == "right"):
self.rotateRight()
time.sleep(1.0)
self.allStop()
return
def stateMachine(self):
print "Starting recording"
prevState = None
self.microphone.startRecording()
time.sleep(1.0)
while True:
if prevState != self.microphone.read():
prevState = self.microphone.read()
self.on_event(self.microphone.read())
time.sleep(1.0)
if (self.state.__str__() == "Idle"):
continue
#self.allStop()
elif (self.state.__str__() == "FollowHumanIdle"):
self.camera.startStream()
self.allStop()
elif (self.state.__str__() == "FollowHuman"):
self.followHuman()
elif (self.state.__str__() == "VoiceControlIdle"):
self.allStop()
elif (self.state.__str__() == "RotateLeft"):
self.rotateLeft()
elif (self.state.__str__() == "RotateRight"):
self.rotateRight()
elif (self.state.__str__() == "QuitApp"):
self.quit()
return
def exit_gracefully(self, signal, frame):
print('Trying to exit gracefully...')
self.quit()
sys.exit(0)
def __init__(self, master):
self.microphone = Microphone()
self.audioToNet = AudioToNet()
self.trafficPrinter = TrafficPrinter()
self.master = master
