This is for audio recognition, a DSP task of XJTU, created by Xinye Wang, Ziyang Tang, Yidong Lu, Qin Zhao and Yixin Chen.
The project includes three subtasks: speech recognition based on time domain analysis techniques, speech recognition based on frequency domain analysis techniques and Content-independent speaker recognition(optional).
The "dataset" folder contains 17 .wav files sampled from 17 people, each pronouncing the ten numbers from 0 to 9.
The "guidance" folder contains material taken from the Internet (mainly the Chinese Internet).
Three tasks corresponds to different folders.