WavFT: Acoustic model finetuning with labelled and unlabelled data

no code implementations • 1 Apr 2022 • Utkarsh Chauhan, Vikas Joshi, Rupesh R. Mehta

Unsupervised and self-supervised learning methods have leveraged unlabelled data to improve the pretrained models.

Self-Supervised Learning

Transfer Learning Approaches for Streaming End-to-End Speech Recognition System

no code implementations • 12 Aug 2020 • Vikas Joshi, Rui Zhao, Rupesh R. Mehta, Kshitiz Kumar, Jinyu Li

Transfer learning (TL) is widely used in conventional hybrid automatic speech recognition (ASR) system, to transfer the knowledge from source to target language.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +2