Scribd
Upload
6 views

2. Synopsis Content

This project focuses on speech-based emotion recognition, classifying emotions into six categories using Mel Frequency Cepstral Coefficients (MFCC) and the CREMA-D dataset with data augmentation techniques. The methodology involves training models using Convolutional Neural Networks (CNN), Support Vector Machine (SVM), and Random Forest (RF) to achieve accurate emotion classification. The system aims to enhance human-machine interactions by enabling machines to recognize emotions in speech, with potential applications in voice assistants and call centers.

Uploaded by

Mariyam tasmiya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
6 views

2. Synopsis Content

This project focuses on speech-based emotion recognition, classifying emotions into six categories using Mel Frequency Cepstral Coefficients (MFCC) and the CREMA-D dataset with data augmentation techniques. The methodology involves training models using Convolutional Neural Networks (CNN), Support Vector Machine (SVM), and Random Forest (RF) to achieve accurate emotion classification. The system aims to enhance human-machine interactions by enabling machines to recognize emotions in speech, with potential applications in voice assistants and call centers.

Uploaded by

Mariyam tasmiya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

ABSTRACT

This project focuses on speech-based emotion recognition, aiming to classify emotions into six
categories: angry, fear, disgust, happy, sad, and neutral. The proposed approach utilizes Mel
Frequency Cepstral Coefficients (MFCC) for feature extraction and employs the CREMA-D dataset,
enhanced with data augmentation techniques such as noise addition and speed variations. The
extracted features undergo normalization, label encoding, or one-hot encoding, depending on the
model used. The dataset is then split into training and testing sets and evaluated using Convolutional
Neural Networks (CNN), Support Vector Machine (SVM), and Random Forest (RF) to achieve
accurate emotion classification.
INTRODUCTION

The human voice is very versatile and carries a multitude of emotions. Emotion in speech carries
extra insight about human actions. Human speech conveys information and context through speech,
tone, pitch and many such characteristics of the human vocal system. As human machine interactions
evolve, there is a need to buttress the outcomes of such interactions by equipping the computer and
machine interfaces with the ability to recognize the emotion of the speaker. Emotions play a vital
role in human communication. In order to extend its role towards the human-machine interaction, itis
desirable for the computers to have some built-in abilities for recognizing the different emotional
states of the user [2,5].Today, a large amount of resources and efforts are being put into the
development of artificial intelligence, and smart machines, all for the primary purpose of simplifying
human life. Research studies have provided evidence that human emotions influence the decision
making process to a certain extent [1-4]. If the machine is able to recognize the underlying emotion
in human speech, it will result in both constructive response and communication. In order to
communicate effectively with people, the systems need to understand the emotions in speech.
Therefore, there is a need to develop machines that can recognize the paralinguistic information like
emotion to have effective clear communication like humans. One important data in paralinguistic
information is Emotion, which is carried along with speech.
LITERATURE SURVEY
METHODOLOGY

The suggested system's MFCC function was utilized to divide audio data into several emotion
categories using artificial neural networks. Neural networks offer the benefit of being able to
categories many sorts of emotions present in variable-length audio inputs in a real-time setting. This
approach presents a reasonable balance between the precision of execution and the complexity of
real-time approaches.
The speech emotion recognition system is performed as a Machine Learning (ML) model. The steps
of operation are similar to any other ML project, with supplementary fine- tuning systems to make
the model function adequately.
The model being generated will acquire from the data contributed to it and all the conclusions and
decisions that a progressed model will produce is supervised data. The secondary action, called as
feature engineering, is a combination of various machine learning assignments that are performed
over the gathered data. These systems approach the various data description and data quality
problems. The third step is often explored the essence of an ML project where an algorithmic based
prototype is generated. This model uses an ML algorithm to determine about the data and instruct
itself to react to any new data it is exhibited to. The ultimate step is to estimate the functioning of the
built model. Very frequently, developers replicate the steps of generating a model and estimating it
to analyze the performance of various algorithms. Measuring outcomes help to choose the suitable
ML algorithm most appropriate to the p Dataset.

SOFTWARE REQUIREMENTS:
 Operating system : Windows 7 Ultimate.

 Coding Language : Python.

 Front-End : Python.

 Back-End : Django-ORM

 Designing : Html, CSS, Java script.

 Data Base : MySQL (WAMP Server).


POSSIBLE OUTCOME

1. Speech Emotion Recognition – The system will detect and classify emotions from speech.
2. Model Performance Analysis – Comparison of different models like CNN, SVM, and
Random Forest.
3. Practical Application – Can be used in voice assistants, call centers, and emotion analysis
tools.
REFERENCES
[1] Al- Sahib Nabeel.K& Azeez.Z Mohammed created “Internet Remote Control Interface for a
Multi-Purpose Mobile Robot”.
[2] Andreasson.J, Gavert.M created “The Vehicle Dynamics Library Overview and Applications
Modelon”.
[3] Cuno Pfister created “Getting started with the Internet of Things”.
[4] Joy Deep Biswas created “Wifi Localization and Navigation for Autonomous Indoor Mobile
Robots”.
[5] Sudha Jamthe created”IoT Disruptions: The Internet of Things - Innovations & Jobs.
[6] “Tom Nguyen ,josh slonaker created “Semi Autonomous wireless control robot”.
[7] M.S.Sreejith, Steffy Joy created “Conceptual design of a wifi and gps based robotic library using
an intelligent system.
[8] C.W.Warren , S.G.Buckley created “Mobile robotic over wifi network using arduino
technology”.

You might also like