RESEARCH

Acoustic & Speech

Investigates various Speech signal processing schemes for acoustic modeling so that more robust speech recognition can be achieved. Our aim is to perform the state-of-art research providing effective means for achieving:

 Real-Time Lip-Synch System

작성자 관리자 날짜 2021-04-08 21:40:25 조회수 66

 Real-Time Lip-Synch System

 

     Contents

1. Introduction

2. Procedure of Real-Time Lip-Synch System

3. Algorithm for Real-Time Lip-Synch System

4. Applications

 

 

1. Introduction         

  1.1 What is the lip-synch system: Activate avatar's lip motion in synch with the speech signal.

  • Real-time lip-synch system: To synchronize the lip movements of the avatar with the audio signal in real time perfectly

 

2. Procedure of Real-Time Lip-Synch System        

  2.1 Screen Shot of the Real-Time Lip-Synch System( Real-Time Speech Input)

  

  2.2 Procedure of Real-Time Lip-Synch System

         

3. Algorithm for Real-Time Lip-Synch System        

  3.1 High performance

  • Head-body-tail (HBT) model : The variation can be effectively modeled
  • The center of the each phoneme, represented by the body model, is a context independent units

  • Context dependency information is incorporated in the Head/Tail models

  3.2 Mixture tying of HBT models

  • Semi-continuous Hidden Markov Model (SCHMM) : Reduction of the parameters

4. Applications        

  • Computer graphic animation lip-synch
  • Avatar chatting
  • Virtual reality system
  • Foreign language pronunciation study
  • In noise environments, lip movement can help listening

 

댓글 (0)

등록된 댓글이 없습니다.
작성 권한이 없습니다.