About me
I am a recent M.Sc. graduate in Computer Software Engineering from the University of Tehran (GPA: 4.0/4.0), with a research focus on conversational AI and speech processing. My master’s thesis, supervised by Dr. Shakery and Dr. Faili, centered on zero-shot text-to-speech models for the Persian language. As part of this work, I led the creation of ParsVoice, a large-scale multi-speaker Persian speech corpus consisting of over 3.5K hours of curated audiobook and podcast recordings, and developed TTS models tailored to Persian. I will be joining the EPFL NLP Lab as a research intern in the upcoming spring.
As a Research and Teaching Assistant at the University of Tehran's NLP Lab and Intelligent Information Systems Lab, I lead research on Persian speech synthesis and medical language models. I have designed comprehensive assignments for Machine Learning Methods in NLP and Digital Speech Processing courses, serving over 200+ students across multiple courses. Additionally, I co-founded SOFIA, where I architected and deployed a multi-platform RAG-based customer service chatbot that has achieved 20,000+ active users on University of Tehran's website.
Before my graduate studies, I earned my Bachelor's degree in Electrical Engineering from Amirkabir University of Technology (AUT), where I worked on facial expression recognition with robotic implementation under the guidance of Dr. Menhaj and Dr. Taheri. I have consistently demonstrated academic excellence, ranking in the top 0.5% in national entrance exams.
Research Interests
- Conversational AI
- Large Language Models
- Multimodal AI
- Controllable Generative Models
- Speech Synthesis & Processing
- Natural Language Processing
Education
- Master of Science in Computer Software Engineering, 2022 -- Present
University of Tehran (GPA: Perfect 4.0/4.0)
Thesis: Zero-Shot Text-to-Speech for Persian Language - Bachelor of Science in Electrical Engineering, 2016 -- 2021
Amirkabir University of Technology
Thesis: Facial Expression Recognition with Robotic Implementation
Publications
- MJ. Ranjbar, Mohammad Amini, Parmis Bathaeyian, Zhengjun Yue, H. Faili, A. Shakery. PARSA-Bench: A Comprehensive Persian Audio-Language Model Benchmark. Under preparation for submission to Interspeech 2026.
- MJ. Ranjbar, H. Faili, A. Shakery. ParsVoice: A Large-Scale Multi-Speaker Persian Speech Corpus for Text-to-Speech Synthesis. Under preparation for submission to Interspeech 2026.
- MJ. Ranjbar, A. Shakery, H. Faili. PersianPunc: Neural Punctuation Restoration for Persian Speech Recognition and Text Processing. Under review at EACL SilkRoadNLP Workshop.
- MJ. Ranjbar, A.H. Sheikh, S. Ranjbar Kalahroodi, S. Karimi, A. Shakery, H. Faili. PersianMedQA: Evaluating Large Language Models on a Persian-English Bilingual Medical Question Answering Benchmark. Under review at LREC 2026.
- MJ. Ranjbar, MB. Menhaj, H. Taheri. Social Robots: An Open-Source Framework for Personal Assistant Robots. Published in ICRoM 2022.