Two-pass Endpoint Detection for Speech Recognition

no code implementations • 17 Jan 2024 • Anirudh Raju, Aparna Khare, Di He, Ilya Sklyar, Long Chen, Sam Alptekin, Viet Anh Trinh, Zhe Zhang, Colin Vaz, Venkatesh Ravichandran, Roland Maas, Ariya Rastrow

Endpoint (EP) detection is a key component of far-field speech recognition systems that assist the user through voice commands.

speech-recognition Speech Recognition

Adaptive Endpointing with Deep Contextual Multi-armed Bandits

no code implementations • 23 Mar 2023 • Do June Min, Andreas Stolcke, Anirudh Raju, Colin Vaz, Di He, Venkatesh Ravichandran, Viet Anh Trinh

In this paper, we aim to provide a solution for adaptive endpointing by proposing an efficient method for choosing an optimal endpointing configuration given utterance-level audio features in an online setting, while avoiding hyperparameter grid-search.

Multi-Armed Bandits

Towards Adapting NMF Dictionaries Using Total Variability Modeling for Noise-Robust Acoustic Features

1 code implementation • 16 Jul 2019 • Kunal Dhawan, Colin Vaz, Ruchir Travadi, Shrikanth Narayanan

We propose an algorithm to extract noise-robust acoustic features from noisy speech.

SAIL: Sentiment Analysis using Semantic Similarity and Contrast Features

no code implementations • SEMEVAL 2014 • Mal, Nikolaos rakis, Michael Falcone, Colin Vaz, Jesse James Bisogni, Alex Potamianos, ros, Shrikanth Narayanan

Language Modelling Semantic Similarity +2