Description

Development of a Voice-Based AI System for Real-Time Emotion Detection in Human Conversations


General Description:
This project focuses on the development and enhancement of a real-time voice emotion recognition AI system—Voice Language Model (VLM). The system analyses live conversations between two interlocutors to detect emotions such as frustration, stress, satisfaction, etc. 
Using advanced voice signal processing and psychological research, the VLM captures subtle vocal cues (e.g., pitch, tempo, tremble) to understand the speaker’s emotional and psychological profile. 
The project aims to improve the accuracy and adaptability of emotion detection models, contributing to the real-world application of emotionally intelligent AI.

This project isn’t just about building smart tech—it’s about making conversations more human. 
VLM helps people understand and respond to the emotions of others in real time, leading to more respectful, empathetic interactions. 

The result? People feel heard and cared for. It’s a step toward a world where technology doesn’t just listen—it understands.

Project Scope:
The objective is to develop a real-time AI model that can detect and classify human emotions based on vocal input alone, using dual-channel recordings. The model should be capable of recognizing at least 8–10 core emotions (based on Plutchik’s Wheel) and track emotional dynamics over time. The project also aims to explore how vocal markers such as jitter, pitch, shimmer, etc., correlate with specific psychological states like stress and fatigue, or joy and excitement. Students will contribute to developing, training and validating the model, working with real datasets, and improving detection accuracy using a combination of traditional signal processing and machine learning.

Student Requirements:
● Excitement about building the tech that makes the world a bit better
● Strong interest in working in the field of applied AI and voice analysis
● Full attendance in weekly progress meetings
● Ability to conduct independent research
● Personal responsibility for assigned tasks and deadlines
● Good teamwork and collaboration skills
● Desire to learn from the industry experts
● Motivation to experience what it’s like to build real solutions in a professional R&D environment before graduation
● Interest in psychology is a plus
Development Tools
Python, Jupiter Notebook, Tensorflow, Git
Deliverables
- GitHub repository containing all source code
- Technical documentation and setup instructions
- Model performance summary (metrics, confusion matrix, etc.)
- Brief report on methodology, key findings, and recommendations for future improvements
Project Coordinators
Veronika Shpareva veronika@voxspec.com Josh (Yegor) Zaslavskiy yz@voxonics.com
Additional Notes
• Each project team may include up to 5 students. • Please include your contact details at the end of the document for student inquiries: - Full name - Email address.

Deliverables for Submission:

One Pager – Project Summary:
One-page overview with problem, solution, team members, and mentor.

GitHub Link – Source Code:
Clean, organized code with clear comments.

YouTube Link – Project Video:
Screen recording explaining the code, app, and site. Upload and share the link in a separate document.

Specification Document (English):
Detailed, professional spec in English. Important for grading.

PowerPoint Presentation:
Includes team info, requirements, screenshots, graphs, and mockups.


For any questions, you may also contact: mayyalih@gmail.com
Real-Time Emotion Detection via Voice AI

AI and Machine Learning

Description

Emphasis in project execution

The project is has cooperation with the industry and combines meeting deadlines while being creative and focused on the task
