We will display a rectangle on top of the face. To avoid flickering of the rectangle, we will show it at it latest known position if the face is not detected.
@@ -89,7 +89,7 @@ Pick the right statistical method for your application. TM_CCOEFF (right), TM_SQ
## 局限性
Template matching is not scale invariant nor is it rotation invariant. It is a very basic and straightforward method where we find the most correlating area. Thus, this method of object detection depends on the kind of application you want to build. For non scale and rotation changing input, this method works great.
Speech recognition can by done using the Python SpeechRecognition module. We make use of the [Google Speech API](https://pythonspot.com/speech-recognition-using-google-speech-api/"Google Speech Recognition in Python") because of it’s great quality.
Various [**APIs and programs are available for text to speech applications**](https://pythonspot.com/speech-engines-with-python-tutorial/"TTS"). Espeak and pyttsx work out of the box but sound very robotic. We decided to go with the Google Text To Speech API, gTTS.
各种 [**API 和程序可用于文本到语音的应用程序**](https://pythonspot.com/speech-engines-with-python-tutorial/)。Espeak 和 pyttsx 开箱即用,但听起来很机器人。我们决定使用 Google 文字到语音 API gTTS。