Intelligent Vocal Training Assistant System

In professional vocal training, a way to evaluate the quality of vocalization is often needed. In order to solve various problems caused by the lack of professional instructors, a vocal training system for detecting the closed state of the vocal cords is proposed. We have proposed a robust vocal training system for the detection of the closed state of the vocal cords on the mobile terminal, which can analyze the closure of the vocal cords of the human body during vocalization, so as to evaluate the vocal cord ability without a professional teacher or professional equipment. In this system, we can use two wearable sensors, vibrating plate and headset, to collect the signals of the human vocal cords, or directly use the microphone of the mobile device to collect. And we use the convolutional neural network to analyze the signals and classify the closed state of the vocal cords. In order to build this system, we constructed a vowel data set classified by degree vocal cords closure, including the wearable sensors and mobile phone microphones we used. We compared the performance of the traditional vocalization pattern classification method and the convolutional neural network method we used to classify the vocal cord closure types on the data set we constructed and the public data set, and finally tested under two noise environments, and the preliminary results proved the usability of our system.