Chandrashekhar Lavania — accepted AI-conference papers

2025

Improving Lip-synchrony in Direct Audio-Visual Speech-to-Speech Translation

Lucas Goncalves, Prashant Mathur, Xing Niu, Chandrashekhar Lavania, Brady Houston, Srikanth Vishnubhotla +2

ICASSP 2025accepted

Audio-Visual Speech-to-Speech Translation (AVS2S) typically prioritizes improving translation quality and naturalness. However, an equally critical aspect in audio-visual content is lip-synchrony—ensuring that the movements of the lips match the spoken content—essential for maintaining realism in du…

Cited by 0 Source ↗Scholar ↗

2024

Perceptual Evaluation of Audio-Visual Synchrony Grounded in Viewers’ Opinion Scores

Lucas Goncalves, Prashant Mathur*, Chandrashekhar Lavania, Metehan Cekic, Marcello Federico, Kyu Han

ECCV 2024poster

"Recent advancements in audio-visual generative modeling have been propelled by progress in deep learning and the availability of data-rich benchmarks. However, the growth is not attributed solely to models and benchmarks. Universally accepted evaluation metrics also play an important role in advanc…

Cited by 1 Source ↗PDF ↗Scholar ↗

2023

Multi-Scale Compositional Constraints for Representation Learning on Videos

Georgios Paraskevopoulos, Chandrashekhar Lavania, Lovish Chum, Shiva Sundaram

ICASSP 2023accepted

Combining simple concepts to form structured thoughts and decomposing complex concepts into their constituents is one key characteristic of human cognition. In this work we extract video representations by combining multi-scale processing with compositional constraints, i.e., we constrain the latent…

Cited by 0 Source ↗Scholar ↗

2022

Enhancing Contrastive Learning with Temporal Cognizance for Audio-Visual Representation Generation

Chandrashekhar Lavania, Shiva Sundaram, Sundararajan Srinivasan, Katrin Kirchhoff

ICASSP 2022accepted

Audio-visual data allows us to leverage different modalities for downstream tasks. The idea being individual streams can complement each other in the given task, thereby resulting in a model with improved performance. In this work, we present our experimental results on action recognition and video…

Cited by 0 Source ↗Scholar ↗

2021

Constrained Robust Submodular Partitioning

Shengjie Wang, Tianyi Zhou, Chandrashekhar Lavania, Jeff Bilmes

NeurIPS 2021spotlight

In the robust submodular partitioning problem, we aim to allocate a set of items into $m$ blocks, so that the evaluation of the minimum block according to a submodular function is maximized. Robust submodular partitioning promotes the diversity of every block in the partition. It has many applicatio…

Cited by 8 Source ↗PDF ↗Scholar ↗

2019

Fixing Mini-batch Sequences with Hierarchical Robust Partitioning

Shengjie Wang, Wenruo Bai, Chandrashekhar Lavania, Jeff Bilmes

AISTATS 2019poster

We propose a general and efficient hierarchical robust partitioning framework to generate a deterministic sequence of mini-batches, one that offers assurances of being high quality, unlike a randomly drawn sequence. We compare our deterministically generated mini-batch sequences to randomly generat…

Cited by 12 Source ↗PDF ↗Scholar ↗

2017

Reducing total latency in online real-time inference and decoding via combined context window and model smoothing latencies

Chandrashekhar Lavania, Jeff A. Bilmes

ICASSP 2017accepted

Real-time low-latency online inference and decoding in sequential probabilistic models are important in many interactive systems, including automatic speech recognition (ASR) and streaming environments. We study total inference latency (TL) in such systems, the additively combined latency of the inh…

Cited by 0 Source ↗Scholar ↗