Improving Confidence Estimation on Out-of-Domain Data for End-to-End Speech Recognition

TL;DR

This paper introduces methods to enhance confidence estimation for end-to-end speech recognition systems when faced with out-of-domain data, using pseudo transcriptions and additional language models, leading to more reliable confidence metrics.

Contribution

It proposes two novel approaches to improve model-based confidence estimators on out-of-domain data without altering the ASR model itself.

Findings

Significant improvement in confidence metrics on OOD datasets

Better calibration of confidence estimators on out-of-domain data

Enhanced data selection reliability using improved confidence estimates

Abstract

As end-to-end automatic speech recognition (ASR) models reach promising performance, various downstream tasks rely on good confidence estimators for these systems. Recent research has shown that model-based confidence estimators have a significant advantage over using the output softmax probabilities. If the input data to the speech recogniser is from mismatched acoustic and linguistic conditions, the ASR performance and the corresponding confidence estimators may exhibit severe degradation. Since confidence models are often trained on the same in-domain data as the ASR, generalising to out-of-domain (OOD) scenarios is challenging. By keeping the ASR model untouched, this paper proposes two approaches to improve the model-based confidence estimators on OOD data: using pseudo transcriptions and an additional OOD language model. With an ASR model trained on LibriSpeech, experiments show that the proposed methods can greatly improve the confidence metrics on TED-LIUM and Switchboard datasets while preserving in-domain performance. Furthermore, the improved confidence estimators are better calibrated on OOD data and can provide a much more reliable criterion for data selection.