Saved in:
Bibliographic Details
Main Authors: Aggarwal, Vaibhav, Nair, Shabari S, Verma, Yash, Jogi, Yash
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.13446
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909502131929088
author Aggarwal, Vaibhav
Nair, Shabari S
Verma, Yash
Jogi, Yash
author_facet Aggarwal, Vaibhav
Nair, Shabari S
Verma, Yash
Jogi, Yash
contents Recent research on word-level confidence estimation for speech recognition systems has primarily focused on lightweight models known as Confidence Estimation Modules (CEMs), which rely on hand-engineered features derived from Automatic Speech Recognition (ASR) outputs. In contrast, we propose a novel end-to-end approach that leverages the ASR model itself (Whisper) to generate word-level confidence scores. Specifically, we introduce a method in which the Whisper model is fine-tuned to produce scalar confidence scores given an audio input and its corresponding hypothesis transcript. Our experiments demonstrate that the fine-tuned Whisper-tiny model, comparable in size to a strong CEM baseline, achieves similar performance on the in-domain dataset and surpasses the CEM baseline on eight out-of-domain datasets, whereas the fine-tuned Whisper-large model consistently outperforms the CEM baseline by a substantial margin across all datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2502_13446
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Adopting Whisper for Confidence Estimation
Aggarwal, Vaibhav
Nair, Shabari S
Verma, Yash
Jogi, Yash
Audio and Speech Processing
Machine Learning
Recent research on word-level confidence estimation for speech recognition systems has primarily focused on lightweight models known as Confidence Estimation Modules (CEMs), which rely on hand-engineered features derived from Automatic Speech Recognition (ASR) outputs. In contrast, we propose a novel end-to-end approach that leverages the ASR model itself (Whisper) to generate word-level confidence scores. Specifically, we introduce a method in which the Whisper model is fine-tuned to produce scalar confidence scores given an audio input and its corresponding hypothesis transcript. Our experiments demonstrate that the fine-tuned Whisper-tiny model, comparable in size to a strong CEM baseline, achieves similar performance on the in-domain dataset and surpasses the CEM baseline on eight out-of-domain datasets, whereas the fine-tuned Whisper-large model consistently outperforms the CEM baseline by a substantial margin across all datasets.
title Adopting Whisper for Confidence Estimation
topic Audio and Speech Processing
Machine Learning
url https://arxiv.org/abs/2502.13446