Scribd
Upload
13 views5 pages

9 Removed

Uploaded by

srichandramounikaankam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
13 views5 pages

9 Removed

Uploaded by

srichandramounikaankam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

ABSTRACT

Speech emotion recognition has become a hot topic and hasbeen developed in the
subject of Human Computer Interaction ( HCI ). Speech recognition (SR) is a field of
technology that involves methods and fields for identifying speech from speech signals.
Different technological; trends withinside the vicinity of the SPMs ( Signal Processing
Methods ), the popularity of the expression has end up probable.Despite the fact that there
has been a significant increase in the field of voice recognition, there are various voice
yields that have been implemented like Alexa and Siri that are based on voice-based
commands. SER (Speech Emotion Recognition) is a research topic that aims to bring
together feelings expressed by the speech signals. Different surveys stated that the
development in sentiment detection made a lot of the networks simpler and the world a
better location for living.

Emotion detection is a difficult problem because the emotion might vary depending
on the situation, culture, and person's face-response, resulting in unclear findings; speech
quantity is insufficient to accurately infer the emotion; and many languages lack a speech
database. SER has also been employed in a variety of applications such as robot interaction,
bank services, digital gaming, and so on.

Different speech emotions such as Happy, Anger, Neutral, and Sad have been
detected or recognised using feature vectors in previous studies.Voice Pitch, MFCC (Mel
Frequency Cepstral Coefficients), and STM were among the feature sets deleted from the
audio data (Short Term Energy). The influence of the rising number of feature sets provides
for the classifier, as diverse methodologies have built on the feature sets. We will develop a
model using an MLPClassifier using the libraries librosa, sound file, and sklearn in this
project. This will be able to detect emotion in audio files. The data will be loaded, features
extracted, and the dataset divided into training and testing sets. The model will then be
trained using an MLPClassifier that has been initialized.

I
LIST OF CONTENTS

ABSTRACT I
IV
LIST OF FIGURES

LIST OF TABLES V

CHAPTER NO TITLE PAGE NO


CHAPTER 1 INTRODUCTION 2
1.1 OBJECTIVE 3
1.2 ABOUT THE ALGORTHIM 4
1.3 PURPOSE 5
1.4 SCOPE 6
LITERATURE REVIEW
CHAPTER 2 8
CHAPTER 3 SRS DOCUMENT 13
3.1 FUNCTIONAL REQURIMENTS 13
NON FUNCTIONAL
3.2 REQURIMENTS 14
MINIMUM SOFTWARE
3.3 REQURIMENTS 15
MINIMUM HARDWARE
3.4 16
REQURIMENTS
CHAPTER 4 ALGORITHM ANALYSIS 18
4.1 EXISTING ALGORTHIM 18
4.2 PROBLEM STATEMENT 18
PROPOSED ALGORTHIM 20

4.3
ADVANTAGES OF PROPOSED
22
ALGORTHIMS
4.4 COST BENEFIT ANALYSIS 22
CHAPTER 5 SOFTWARE DESCRIPITION 25
5.1 GOOGLE COLAB 25
5.2 STREAMILT 25
5.3 NGROK 26
5.4 LIBROSA 26
PIP INSTALL LIBROS--USER 26
IMPORT LIBROSA AS LIB 26
LIBROSA.CORE.LOAD() 26
SKLEARN 27
PIP INSTALL SCIKIT -LEARN 27
5.5 NUMPY 28

II
CHAPTER 6 SYSTEM DESIGN 30
6.1 INTRODUCTION TO UML 30
GOALS OF UML 30
6.2 UML DIAGRAM 33
6.3 IMPLEMENTATION 34
6.4 METHODOLOGY 37
CHAPTER 7 DEVELOPMENT 39
7.1 DATASET USED 39
CHAPTER 8 TESTING 43
8.1 INTRODUCTION TO TESTING 43
IMPORTANCE OF TESTING 43
8.2 BENEFITS OF TESTING 44
*DIFFERENT TYPES OF
TESTING 45
BLACK BOX TESTING 47
WHITE BOX TESTING 48
8.3 TEST CASE 52
RESULT 56
CONCLUSIONS 61
FUTURE SCOPE 62
REFERENCES 63

III
LIST OF FIGURES

FIG NO TITLE PAGE NO

FIG 1.2 MLP NEURAL NETWORK 04

FIG 1.2 WORK FLOW OF MLP CLASSIFER 05

FIG 3 General steps for Speech Recognition System 13


(SRS)
FIG 4.3 PROPOSED ALGORTHIM 20

FIG 4.3 MLP CLASSIFER 22

FIG 5.2 STREAMILT 24

FIG 5.4 FEATURE EXTRACTION 26

FIG 6.1 A BUILDING BLOCKS IN UML 30

FIG 6.1 B SYSTEM ARCHITECTURE 31

FIG 6.2A UML DIAGRAM 32

FIG 6.2B SEQUENCE DIAGRAM 33

FIG 6.4 PICTORIAL REPRESENTATION OF SPEECH 36


FEATURE EXTRACTION

FIG 8.2A BLACK BOX TESTING 47

FIG 8.2B PATH TESTING 48

FIG 8.2C WHITE BOX TESTING 49

FIG 9A SHOWING WAVEPLOT FOR FEAR 54

FIG 9B SHOWING SPECTROGRAM FOR FEAR 55

FIG 9C SHOWING WAVEFROM AND 55


SPECTROGRAM FOR ANGRY 56

FIG 9D SHOWING WAVEFROM AND 57


SPECTROGRAM FOR HAPPY

IV
LIST OF TABLES

TABLE NO TABLES NAMES PAGE NO


TABLE 1 ML classifier 9
Literature Survey of
TABLE 2 Emotion Recognition from 11
Speech

You might also like