Start Recording

LitmusBox_StartRecording.py

def start_recording(self):
	self.recording = 1
	while self.recording == 1:
	  # Find haar cascade to draw bounding box around face
	  frame = self.vs.read()

	  self.videoFrameStore.append(frame)

	  # fps stuff
	  self.out.write(frame)
	  self.fps.update()

	  # audio stuff
	  data = self.stream.read(self.chunk, exception_on_overflow=False)
	  self.audioFrames.append(data)
	  self.win.update()


	for currentVideoFrame in self.videoFrameStore:
	  # convert to gray and get face position
	  gray = self.cv2.cvtColor(currentVideoFrame, cv2.COLOR_BGR2GRAY)
	  faces = self.facecasc.detectMultiScale(
		  gray, scaleFactor=1.3, minNeighbors=5)

	  for (x, y, w, h) in faces:
		  roi_gray = gray[y:y + h, x:x + w]
		  cropped_img = np.expand_dims(np.expand_dims(
			  cv2.resize(roi_gray, (48, 48)), -1), 0)

		  prediction = self.model.predict(cropped_img)
		  maxindex = int(np.argmax(prediction))
		  # increment score
		  emotion = self.emotion_dict[maxindex]
		  self.emotion_score_dict[emotion] += 1

	# write to audio file
	self.stream.stop_stream()
	self.stream.close()
	self.audio.terminate()
	self.fps.stop()

	# save the audio frames as .wav file
	wavefile = self.wave.open(self.wav_output_filename, 'wb')
	wavefile.setnchannels(self.chans)
	wavefile.setsampwidth(self.audio.get_sample_size(self.form_1))
	wavefile.setframerate(self.samp_rate)
	wavefile.writeframes(b''.join(self.audioFrames))
	wavefile.close()

	# write final emotion score to txt file
	with open(self.dateTimeString + '.txt', 'w') as file:
	  file.write(json.dumps(self.emotion_score_dict))

	print("[INFO] elasped time: {:.2f}".format(self.fps.elapsed()))
	print("[INFO] approx. FPS: {:.2f}".format(self.fps.fps()))

	# cap.release()
	self.out.release()
	self.cv2.destroyAllWindows()
	self.vs.stop()