Soujanya Poria

This repo contains implementation of different architectures for emotion recognition in conversations.

A family of diffusion models for text-to-audio generation.

MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversation

This repository contains various models targetting multimodal representation learning, multimodal fusion for downstream tasks such as multimodal sentiment analysis.

[ICLR 2026] TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching

This repository contains code to quantitatively evaluate instruction-tuned models such as Alpaca and Flan-T5 on held-out tasks.