mgvt

Modular architecture for an AI-based 3D Virtual Teacher (Bachelor's thesis 2025)

https://github.com/michele-giordano/mgvt

Science Score: 67.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
    Found CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
    Found 2 DOI reference(s) in README
  • Academic publication links
    Links to: zenodo.org
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (10.0%) to scientific vocabulary

Keywords

ai architecture llm mgvt raef tts verbamanent virtual-teacher
Last synced: 8 months ago · JSON representation ·

Repository

Modular architecture for an AI-based 3D Virtual Teacher (Bachelor's thesis 2025)

Basic Info
  • Host: GitHub
  • Owner: michele-giordano
  • License: other
  • Default Branch: main
  • Homepage:
  • Size: 39.1 KB
Statistics
  • Stars: 0
  • Watchers: 0
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Topics
ai architecture llm mgvt raef tts verbamanent virtual-teacher
Created 10 months ago · Last pushed 10 months ago
Metadata Files
Readme License Citation

README.md

MGVT - Multimodal Generative Virtual Teacher

License: GPL v3 DOI

This repository hosts the core materials and code developed for the Bachelor’s thesis entitled "Design and Development of an AI-based Virtual Teacher for Education", completed by the author in 2025 as part of a Bachelor’s Degree in Computer and Software Engineering.

Overview

MGVT is an application that adopts the architectural principles of the MARE framework to implement an advanced AI-driven Virtual Teacher. It combines natural language processing, real-time speech synthesis, and 3D avatar rendering in a modular design.

This project also led to the creation of several specialized components, including:

  • Live functional demo: A Unity3D application featuring real-time lipsync and a distributed asynchronous pipeline.
  • VerbaManent: A customized fork of Coqui-TTS, restructured for improved modularity, phoneme alignment, and performance on consumer hardware.
  • MARE framework: A modular evolution of the original RAEF architecture described in the author's thesis, built on virtual roles (Writer, Reviewer, Dubber, Courier) to optimize text-to-speech generation in educational scenarios.

License

This project is released under the GNU General Public License v3 (GPL-3.0),
to promote open-source and academic research use.
Any proprietary, closed-source, or commercial use requires prior written authorization.

See the LICENSE file for complete terms.

For commercial licensing or special agreements, please contact Michele Giordano.

References and related work

  • VerbaManent
  • MARE framework
  • Upcoming publications will further describe the MGVT framework and its potential applications in education and beyond.

For questions or collaboration opportunities, please contact:
giordano.michele@gmail.com

Owner

  • Name: Michele Giordano
  • Login: michele-giordano
  • Kind: user

Citation (CITATION.cff)

cff-version: 1.2.0
message: "If you use MGVT or any part of this work, please cite it as below."
title: "MGVT - Michele Giordano’s Virtual Teacher"
authors:
  - family-names: Giordano
    given-names: Michele
    orcid: https://orcid.org/0009-0008-9375-5424
date-released: 2025-06-11
version: 0.1.1
license: GPL-3.0-only
repository-code: https://github.com/michele-giordano/mgvt
doi: 10.5281/zenodo.15769233
keywords:
  - AI
  - virtual teacher
  - speech synthesis
  - educational technology
  - lipsync
  - research

GitHub Events

Total
  • Release event: 1
  • Push event: 6
  • Create event: 1
Last Year
  • Release event: 1
  • Push event: 6
  • Create event: 1