Low-resource speech recognition and dialect identification of Irish in a multi-task framework

TL;DR

This paper presents a multi-task hybrid CTC/Attention model for low-resource Irish speech recognition and dialect identification, achieving significant improvements over existing models and approaching state-of-the-art performance.

Contribution

It introduces a multi-task training framework with Intermediate CTC and compares Conformer and E-branchformer encoders for Irish low-resource speech tasks.

Findings

10.8% relative improvement in dialect identification accuracy

WER approaching TDNN-HMM performance

Multi-task approach is promising for low-resource Irish speech tasks

Abstract

This paper explores the use of Hybrid CTC/Attention encoder-decoder models trained with Intermediate CTC (InterCTC) for Irish (Gaelic) low-resource speech recognition (ASR) and dialect identification (DID). Results are compared to the current best performing models trained for ASR (TDNN-HMM) and DID (ECAPA-TDNN). An optimal InterCTC setting is initially established using a Conformer encoder. This setting is then used to train a model with an E-branchformer encoder and the performance of both architectures are compared. A multi-task fine-tuning approach is adopted for language model (LM) shallow fusion. The experiments yielded an improvement in DID accuracy of 10.8% relative to a baseline ECAPA-TDNN, and WER performance approaching the TDNN-HMM model. This multi-task approach emerges as a promising strategy for Irish low-resource ASR and DID.