Speech Recognition(Fall 2021)

Administrative Matters

Instructor: Dr. Ying Shen (yingshen@tongji.edu.cn)

Evaluation: assignments (40%), project (50%), participation (10%)

Latest Notices

Lecture Slides

Slides

Related Materials

Introduction

Speech signal analysis

Assignment 1 due date:

DFT I

DFT II

HMM

Assignment 2 due date:

HMM single Gaussian for isolated words apply EM algorithm (MATLAB codes)

GMM

Assignment 3 due date:

CDHMM

Lab 1

Lab 2

Lab 3

Lab 4

Kaldi toolkit

yesno dataset

data_thchs30

resource

Windows下安装kaldi

Assignments

Notes:

1. Compress all files into a .zip file whose name is composed of student name and ID. (such as "ID_name_assignment1.zip")

2. Plagiarism is forbidden and resubmission will not be accepted.

3. All the documents you hand in, including comments in the source codes, should be in English.

4. Submit your solutions to canvas

Final Projects

Notes:

1. Compress all files into a .rar or .zip file whose name is composed of student name and ID (such as "ID_name_project.zip").

2. All the documents you hand in should be in English.

Requirement details for the program and the report:

Project contents

Program (25 points)

Report (35 points)

Marking

Program:

Origninality of the selected topic or applied method (published since 2010) (5');

Performance (8')

Complexity(7')

Report:

1. (5'); 2. (8'); 3.(5'); 4. (7'); 5. (3'); Clarity (2')

Main References

《语音识别 原理与应用》

洪青阳 李琳

电子工业出版社

Other Related Materials

Spoken language processing: A guide to theory, algorithm and system development

Xuedong Huang, Alex Acero, Hsiao-wuen Hon

Kaldi语音识别实践

陈果果 都家宇 那兴宇 张俊博

电子工业出版社

Created on: Sep. 11, 2018

Last updated on: Dec. 25, 2019