Ishan Biyani

Research Engineer - GenAI (Speech, Multimodal, LLMs) | Diffusion, S2ST, TTS, VC | INTERSPEECH, AAAI

Role
Research Consultant at Sony Research India
Location
Bhilwara, RJ, IN
LinkedIn followers
500 followers
Research & DevelopmentView LinkedIn profile

About Ishan Biyani

Research Engineer working on Speech & Multimodal Generative AI, with hands-on experience building research-grade systems for Speech-to-Speech Translation (S2ST), Text-to-Speech (TTS), Emotional Voice Conversion (EVC), Diffusion models, and LLM/SLM.I currently work with Sony Research India, where I develop and evaluate end-to-end speech generation, voice conversion and translation pipelines, focusing on controllable emotional speech, speaker similarity, and low-resource multilingual settings (English–Hindi / Indic languages). My work bridges applied research and production-oriented GenAI systems, with rigorous objective and perceptual evaluation (MOS, UT-MOS, WER, speaker similarity).I have contributed to peer-reviewed research at INTERSPEECH and and previously worked as a Research Intern at IIT Bombay, developing multilingual OCR. My background includes a Master’s in Machine Learning & AI and training in both academic and industrial research environments.Core interests:• Speech-to-Speech Translation (S2ST)• TTS & Voice Conversion• Diffusion models • Emotional & controllable speech generation• Speech-Language & Multimodal models• LLM-based pipelines• Low-resource and Indic language modelingI’m actively interested in Applied Scientist / Research Engineer roles on teams working at the intersection of Speech, Multimodal AI, and Large Language Models, where research ideas translate into real-world GenAI systems.

Experience

  1. Research Consultant

    Sony Research India

    Aug 2024 — Present · Bengaluru, IN

    Developing a Direct S2ST model for English→Hindi using Omni-LM and streaming TTS• Delivered a production-ready voice de-aging pipeline pitch-shift modules, utilized in TV serials.• Built diffusion-driven VC model with emotion intensity control.• Generated 100+ hours of synthetic bilingual speech data to improve low-resource S2ST performance• First-author INTERSPEECH 2025, co-author 2025

Education

  • International Institute of Information Technology Bangalore

    Executive PG in Machine Learning and AI, Data science

  • Liverpool John Moores University

    Msc in ML and AI, Artificial Intelligence

  • Manipal University Jaipur

    Bachelor of Technology - BTech, computer and communication

    2016 — 2020

Find verified contacts for anyone on LinkedIn

Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.

Free plan included · No credit card required

This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.

Ishan Biyani — Research Consultant at Sony Research India in Bhilwara, RJ, IN | Unifers