CHSER: A Dataset and Case Study on Generative Speech Error Correction for Child ASR

TL;DR

This paper introduces CHSER, a large dataset for generative speech error correction in child ASR, demonstrating significant WER reductions and analyzing the challenges specific to child speech correction.

Contribution

The paper presents the first large-scale dataset for child speech error correction and evaluates its effectiveness in improving child ASR accuracy.

Findings

Up to 28.5% relative WER reduction with zero-shot correction.

13.3% WER reduction when fine-tuning on the dataset.

GenSEC improves substitution and deletion errors but struggles with insertions.

Abstract

Automatic Speech Recognition (ASR) systems struggle with child speech due to its distinct acoustic and linguistic variability and limited availability of child speech datasets, leading to high transcription error rates. While ASR error correction (AEC) methods have improved adult speech transcription, their effectiveness on child speech remains largely unexplored. To address this, we introduce CHSER, a Generative Speech Error Correction (GenSEC) dataset for child speech, comprising 200K hypothesis-transcription pairs spanning diverse age groups and speaking styles. Results demonstrate that fine-tuning on the CHSER dataset achieves up to a 28.5% relative WER reduction in a zero-shot setting and a 13.3% reduction when applied to fine-tuned ASR systems. Additionally, our error analysis reveals that while GenSEC improves substitution and deletion errors, it struggles with insertions and child-specific disfluencies. These findings highlight the potential of GenSEC for improving child ASR.