NETAJI SUBHAS INSTITUTE OF TECHNOLOGY, BIHTA, PATNA 1

PROJECT
ON
VOICE RECOGNITION SYSTEM
GROUP-4

NIDHI RANI (161028)

SHRISTI (161030)
SANSU KUMARI (161036)
PRIYANKA KUMARI (161046)

Dr. SHRADDHA PANDIT Mrs. SARITA CHAUDHARY

(PROJECT GUIDE) (H.O.D)
An ISO 9001:2008 Certified Institute
Approved by AICTE, New Delhi
Affiliated to Aryabhatta Knowledge University, Patna, BIHAR
 2
Overview
• Introduction
• Objective
• Scope
• Hardware and software Requirements
• Existing Models
• Proposed models
• Conclusion
• References
 3

VOICE RECOGNITION
SYSTEM
• the technology by which sounds, words

INTRODUCTION
REQUIREMENTS
PROPOSED

OBJECTIVE
EXISITING

or phrases spoken by humans are

SCOPE

WHY?
MODEL
MODEL

converted into electrical signals, and

these signals are transformed into
coding patterns [1]
 4

VOICE RECOGNITION
SYSTEM
• Mechanism

INTRODUCTION
REQUIREMENTS
PROPOSED

Signal

OBJECTIVE
Speaker
EXISITING

SCOPE
processing and

WHY?
MODEL
MODEL

Modeling
feature Analysis

Classify input
speech based on Speaker profile
existing profiles Database

Output
[2]
 5

OBJECTIVE
• to make this software way more secure than
before
• earlier this system does not provide security

REQUIREMENTS
PROPOSED

Ex:

OBJECTIVE
EXISITING

SCOPE
MODEL
MODEL

As we know the Google assistant

sometimes does not recognise our voice
and gets unlocked by some other
unauthorised user ,so our main focus will
be to make it way more recognisable and
secured to the authorised user.
 6

SCOPES
• Banking
customers can use voice a
REQUIREMENTS

authentication to operate bank

OBJECTIVE
EXISITING

lockers

SCOPE
MODEL
MODEL

• In-app Assistance
Buying products and services with
the sound of your voice

• Solving Crime
Investigating criminals

[2]
 7

HARDWARE AND SOFTWARE

REQUIREMENTS
Hardware:

Microphone

REQUIREMENTS
PROPOSED

OBJECTIVE
EXISITING

Laptop
MODEL

SCOPE
MODEL

Operating System
Pycharm
Xampp

Software:

Python
Machine Learning
Artificial Intelligence
MySQL
 8

EXISTING MODELS
• Google Assistant
• Siri

CATEGORIZATION
•

EVOLUTION?
Amazon Echo

EXISITING

WHY?
MODEL
• Cortana
 9

PROPOSED MODEL
• make this software way more secure than before
because earlier this system does not provide

CATEGORIZATION
security. As we know the Google assistant

EVOLUTION?
PROPOSED

EXISITING
MODEL

WHY?
sometimes does not recognise our voice and gets

MODEL
unlocked by some other unauthorised user ,so our
main focus will be to make it way more
recognisable and secured to the authorised user.
 10

Conclusion
We hope the information we wanted to
convey was easy to understand. The main
motive of the presentation was to spread
some light on the following topics:
• Introduction
• Objective
• Scope
• Hardware and software Requirements
• Existing Models
• Proposed models
 11

REFERENCES:
[1] David Amos, "The Ultimate Guide To Speech Recognition With Python",
[Link]/python-speech-recognition
[2] International journal of advanced research in Computer Science-
[Link]
[3] Jianliang Meng, Junwei Zhang,Haoquan Zhao, "Overview of the Speech
Recognition Technology", Fourth International Conference on Computational and
Information Sciences 2012
 12

THANK YOU