Mohammad Javad Ranjbar

محمدجواد رنجبر

mohammad.ranjbar@epfl.ch  ·  GitHub  ·  LinkedIn  ·  Google Scholar  ·  Hugging Face

Portrait of Mohammad Javad Ranjbar
Currently

Researcher at the EPFL NLP Lab in Lausanne, working with Antoine Bosselut on multilingual audio understanding for Apertus 2, Switzerland's fully open multilingual LLM. Also a research assistant at the BIIC Lab at National Tsing Hua University in Taiwan, working with Chi-Chun Lee on affective speech foundation models across cultures.

I work on multilingual and low-resource NLP, mostly around speech. My research measures how far synthetic (TTS) speech can substitute for real data in low-resource languages, and builds the corpora and benchmarks that Persian NLP has been missing — including ParsVoice, the largest open Persian speech corpus (4,000+ hours), created for my M.Sc. thesis at the University of Tehran (GPA 4.0/4.0) under Azadeh Shakery and Heshaam Faili.

Hear my work بشنوید

Zero-shot voice cloning for Persian: give the model a few seconds of a voice it has never heard, plus a sentence — it speaks the sentence in that voice. This line was never recorded by a human — it's my thesis model, cloning a voice from the ParsVoice demo:

صدای خنده بچه‌ها در تمام کوچه پیچیده بود ▶ press to listen

"The sound of children's laughter echoed throughout the alley."

That covers the generation side. On the understanding side, I built PARSA-Bench (Findings of EACL 2027) — 16 audio-language tasks for Persian, 10 of them new, from speech to classical poetry metre — and at EPFL I'm now bringing audio and text together inside one open multilingual LLM.

Experience تجربه کاری

EPFL NLP Lab · Researcher
Mar 2026 – present

Swiss AI Initiative: adding multilingual audio understanding to Apertus 2. Studying how far synthetic (TTS) speech data can stand in for real speech when pretraining multilingual speech-language models, across six languages, including low-resource ones.

National Tsing Hua University, BIIC Lab · Research Assistant
Jan 2026 – present

With Chi-Chun Lee, on a Taiwan–Germany collaborative project on affective speech foundation models across cultures.

University of Tehran, IIS Lab · Research & Teaching Assistant
Jan 2023 – Mar 2026

Led research on Persian speech synthesis and medical LLMs; chief TA for ML Methods in NLP and Digital Speech Processing; TA for five more courses serving 200+ students.

Sofia · Co-founder & NLP Engineer
Aug 2023 – Aug 2024

Multi-platform RAG customer-service chatbot (Telegram, WhatsApp, web); 20,000+ active users on the University of Tehran website.

Neuromatch Summer School · Teaching Assistant
Summers 2024 – 2026

Mentored international students across the NLP, neuroscience and computer vision courses.

Amirkabir University of Technology & FIRA Academy · Instructor, TA & RA
2020 – 2022

Taught image processing and ML; developed the SWIMBot platform for Diginext; research on facial emotion recognition.

Selected Publications مقاله‌ها

All publications →
How Far Can Synthetic Speech Take Us? Synthetic Data Recipes for Pretraining Multilingual Speech Language Models
M. J. Ranjbar Kalahroodi*, A. K. Tarun*, C. I. Meister, A. Bosselut
Under review (ICLR 2027) · *equal contribution
ParsVoice: A Large-Scale Multi-Speaker Persian Speech Corpus for Text-to-Speech Synthesis
M. J. Ranjbar Kalahroodi, A. Shakery, H. Faili
EMNLP 2026 (Main) · anthology · arXiv · demo · dataset · model · code
PARSA-Bench: A Comprehensive Persian Audio-Language Model Benchmark
M. J. Ranjbar Kalahroodi*, M. Amini*, P. Bathayan, H. Faili, A. Shakery
Findings of EACL 2027 · arXiv · project page · *equal contribution
PersianMedQA: Evaluating Large Language Models on a Persian-English Bilingual Medical Question Answering Benchmark
M. J. Ranjbar Kalahroodi, A. Sheikholselami, S. Karimi Arpanahi, S. Ranjbar Kalahroodi, H. Faili, A. Shakery
LREC 2026 · anthology · arXiv · project page
PersianPunc: A Large-Scale Dataset and BERT-Based Approach for Persian Punctuation Restoration
M. J. Ranjbar Kalahroodi, H. Faili, A. Shakery
SilkRoadNLP @ EACL 2026 · anthology · arXiv · project page

Selected Projects پروژه‌ها

All projects →

Interactive demo of zero-shot voice cloning for Persian, built on the ParsVoice corpus.

Low-cost robot with real-time facial emotion recognition, gesture responses, and speech commands.

Multi-platform chatbot with a custom RAG pipeline; 20,000+ active users at the University of Tehran.

Education تحصیلات

University of Tehran · M.Sc. Computer Engineering
2022 – 2025

GPA 4.0/4.0 · Thesis: Zero-Shot Text-to-Speech for the Persian Language, advised by Azadeh Shakery and Heshaam Faili.

Amirkabir University of Technology · B.Sc. Electrical Engineering
2016 – 2021

Thesis: Facial Expression Recognition with Robotic Implementation, advised by M. B. Menhaj and H. Taheri.

Honors & Awards افتخارات

Ranked 76th of ~15,000 · top 0.5%
2022

National M.Sc. Entrance Exam, Computer Engineering

Ranked 35th of ~2,000 · top 2%
2022

National M.Sc. Entrance Exam, Computer Science

Ranked 411th of ~163,000 · top 0.25%
2016

National University Entrance Exam (Math & Physics)

Invited Talks سخنرانی‌ها

Aug 2026

Audio and Text Understanding for Low-Resource Languages

Jul 2026

Audio and Text Understanding for Low-Resource Languages