presentation template

5
Copyright 2013-2014 Audio and Speech Processing Microproject Speech Recognition using MATLAB Submitted by: Shubham Mittal Roll no.101306160 Shubham Jindal Roll no.101306159

Upload: utkarsh-mittal

Post on 13-Apr-2016

5 views

Category:

Documents


0 download

DESCRIPTION

Audio and Speech

TRANSCRIPT

Page 1: Presentation Template

Copy

right

201

3-20

14

Audio and Speech Processing Microproject

Speech Recognition using MATLAB

Submitted by:Shubham Mittal

Roll no.101306160Shubham Jindal

Roll no.101306159

Page 2: Presentation Template

Copy

right

201

3-20

14

Thursday, April 27, 2023THAPAR UNIVERSITY, PATIALA 2

What is Speech Recognition?

Speech recognition can be defined as the independent, computer‐driven transcription of spoken language into readable text in real time. In a nutshell, SR is technology that allows a computer to identify the words that a person speaks into a microphone or telephone. Having a machine to understand fluently spoken speech has driven speech research for more than 50 years.   Although SR technology is not yet at the point where machines understand all speech, in any acoustic environment, or by any person,  it is used on a day‐to‐day basis in a number of applications and services.

Page 3: Presentation Template

Copy

right

201

3-20

14

Thursday, April 27, 2023THAPAR UNIVERSITY, PATIALA 3

What are the applications of Speech Recognition?

In-car system: like gps navigation and voice controlled mobile etc.Military use: helicopters, stealth fighter planes etc.EducationAutomatic translationRoboticsInteractive voice response: like Siri or CortanaHands-free computingHealth: for physically disabled personsTranscription

Page 4: Presentation Template

Copy

right

201

3-20

14

Thursday, April 27, 2023THAPAR UNIVERSITY, PATIALA 4

Approach?

There are numerous ways in which we can make a speech recognition system. But in this approach we utilise the amplitude of speech signals.First we have a sample sound of a word(Hello) which we read in MATLAB using waveread. Then we record a new voice in realtime and then compare the amplitudes of both in order to decide which word is spoken by the speaker.In order to accomplish this we have taken a threshold value by hit and trial of numerous samples and we take the difference of the maximum amplitude and minimum amplitude above this threshold.

Page 5: Presentation Template

Copy

right

201

3-20

14

Thursday, April 27, 2023THAPAR UNIVERSITY, PATIALA 5

Thankyou