click here to download project abstract
ABSTRACT
Even though artificial intelligence has made tremendous meticulous outcome,
there is a gap and large thirst for speech emotion recognition in affective
computing. Since machines don’t understand the human emotions, natural human
computer interaction is struggling a lot to strive and compete with today’s market.
Survey of past election results and prediction have become crucial and out-dated
since the surveys could never clearly depicts the real mental state of an individual.
Since the wealth and familiarity of the political people has big title role in many
surveys, it is really not trustworthy. No surveys can really understand the truth in
the heart of the common people. So, a novel approach is developed for day to day
physical and psychological behavior of humans which is influenced by emotions.
Other than behavioral expressions, emotion recognition from speech signal is a
challenging modality. For the implementation, we take at least 100 audio signals of
different area, village, town, district, community, religion and privileged people and
we use a nice questionnaire section (20 questions) that could really record the real
emotion of the people. MFCC (Mel-Frequency Cepstral Coefficient) feature is
extracted and classified with Decision Tree, Random Forest and CNN
(Convolutional Neural Network) to give the eight classes of emotions namely
Neutral, Disgust, Calm, Angry, Happy, Sad, Fearful and Surprise. Any traditional
algorithm can be compared to determine the performance of speech emotion
recognition.
vi
TABLE OF CONTENTS
SL.NO
TITLE
ABSTRACT
PAGE.NO
v
LIST OF FIGURES
LIST OF ABBREVIATIONS
vii
viii
1
INTRODUCTION
1
1.1 About Speech Emotion Recognition
1
1.2 Classification of emotions
1
1.3 Existing System
2
1.4 Proposed System
2
1.5 System Requirements
3
2
LITERATURE SURVEY
4
3
METHODOLOGY
8
3.1 Aim
8
3.2 Architecture
8
3.3 Pre-processing
9
3.3.1 Kalman Filter
10
3.4 Feature Extraction
10
3.5 Classification of Emotions
12
3.5.1 Decision Tree
12
3.5.2 Random Forest
14
3.5.3 CNN
15
4
Results and Discussion
18
4.1 Objective of the project
18
4.2 Speech Emotion Recognition
18
5
Conclusion and Future work
23
5.1 Conclusion
24
5.2 Future work
24
References
25
Appendix
27
A. Source Code
27
B. Plagiarism report and paper
33
vii
FIG.NO
LIST OF FIGURES
TITLE
PAGE.NO
3.1
Block Diagram of Speech Emotion Recognition
9
3.2
Enhanced speech
10
3.3.1
Plotting .wav form of audio signal
11
3.3.2
MFCC Feature
11
4.1
Classification of Emotions using Decision Tree
18
4.2
Classification of Emotions using Random Forest
19
4.3
Classification of Emotions using CNN
19
4.4
Confusion matrix
20
4.5
Model loss and Model Accuracy graph
20
4.6
Precision Scores
21
4.7
Recall Values
21
4.8
F1 scores
22
4.9
Comparison of accuracy
22
4.10
Pie chart of classified Emotions
23
