You are on page 1of 13

Speech Recognition System

Project Guide:
Mrs. Rashmi Sharma

Team Members:

Hemanshi Gupta Anukriti Kunwar Ashu Chaudhary


(0703210037) (0703210018) (0703210022)
This SRS provides a complete description of all the
functions and specifications of the Speech
Recognition System. The expected audience of this
project is the faculty and students of ABES. This
document will also serve as a reference for the
developers who will code this system i.e. the team
members of this project.

2
The Speech Recognition System is designed to
work on stand alone systems that satisfy the basic
requirements. It allows users to enter their specific
details which are saved on the hard disk of that
system. This document is the only one that
describes the requirements of the system. It is
meant for use by the developers and will be the
basis for validating the final delivered system.

3
This system will interface with a microphone
and a speaker. The microphone will be
configured to work with the software with the
system.

4
• System Input : Username, Password and 5 voice
samples of a word “hello” uttered at different instances
of time by each user.

• Constraints : Username and password should be text


type. Minimum length for Password is 6 characters.

5
• System Output : If both Text Validation & Speech
Authentication functions are successful for a user then
he’s a valid user, otherwise invalidate the user if he fails
in any of the above functions.

• Constraints : Voice samples saved as .wav file format.


Due to noise, a case may arise when even a valid user is
failed to acknowledged by the system in terms of Speech
Recognition. So, a threshold value is kept that must be
matched in order to be authenticated as a valid user.
6
1) Text Validation
A new user registers by clicking on the “New User
Registration” link. The user enters all its personal
details including username, password (which will be
saved in a .dat file corresponding to that user) and
moves on to the next step of Speech processing,
where samples of a new user are stored for the first
time as a .wav sound file & get stored in the .dat file.

7
Functions (contd.)
2) Speech Authentication : New user records his voice
samples by clicking on the “Record” button and ends the
process by clicking either the “Stop” button or the process
ends in 6 seconds itself (whichever comes earlier). Five
voice samples of the user are taken and stored in the data
repository corresponding to that user. Registration process
ends here. Now, when a user is directed to Speech
Authentication page after being successful in text
validation, he utter “hello” that is matched with those
already stored in data repository .If match occurs, the user
is authenticated, else the system invalidates the user.
8
Speech Processing

9
Use Case Diagram

10
Forgot Password Module

11
New User Module

12
Speech Processing Module

13

You might also like