mohammad.ranjbar@epfl.ch · GitHub · LinkedIn · Google Scholar · Hugging Face
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.
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."
Full demo · dataset (4,000+ hours) · model · code · paper (EMNLP 2026)
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.
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.
With Chi-Chun Lee, on a Taiwan–Germany collaborative project on affective speech foundation models across cultures.
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.
Multi-platform RAG customer-service chatbot (Telegram, WhatsApp, web); 20,000+ active users on the University of Tehran website.
Mentored international students across the NLP, neuroscience and computer vision courses.
Taught image processing and ML; developed the SWIMBot platform for Diginext; research on facial emotion recognition.
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.
GPA 4.0/4.0 · Thesis: Zero-Shot Text-to-Speech for the Persian Language, advised by Azadeh Shakery and Heshaam Faili.
Thesis: Facial Expression Recognition with Robotic Implementation, advised by M. B. Menhaj and H. Taheri.
National M.Sc. Entrance Exam, Computer Engineering
National M.Sc. Entrance Exam, Computer Science
National University Entrance Exam (Math & Physics)
Audio and Text Understanding for Low-Resource Languages
Audio and Text Understanding for Low-Resource Languages