Saved in:
Bibliographic Details
Main Authors: Smith, Griffin Dietz, Yee, Dianna, Chen, Jennifer King, Findlater, Leah
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.23627
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908384305872896
author Smith, Griffin Dietz
Yee, Dianna
Chen, Jennifer King
Findlater, Leah
author_facet Smith, Griffin Dietz
Yee, Dianna
Chen, Jennifer King
Findlater, Leah
contents Identifying mistakes (i.e., miscues) made while reading aloud is commonly approached post-hoc by comparing automatic speech recognition (ASR) transcriptions to the target reading text. However, post-hoc methods perform poorly when ASR inaccurately transcribes verbatim speech. To improve on current methods for reading error annotation, we propose a novel end-to-end architecture that incorporates the target reading text via prompting and is trained for both improved verbatim transcription and direct miscue detection. Our contributions include: first, demonstrating that incorporating reading text through prompting benefits verbatim transcription performance over fine-tuning, and second, showing that it is feasible to augment speech recognition tasks for end-to-end miscue detection. We conducted two case studies -- children's read-aloud and adult atypical speech -- and found that our proposed strategies improve verbatim transcription and miscue detection compared to current state-of-the-art.
format Preprint
id arxiv_https___arxiv_org_abs_2505_23627
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Prompting Whisper for Improved Verbatim Transcription and End-to-end Miscue Detection
Smith, Griffin Dietz
Yee, Dianna
Chen, Jennifer King
Findlater, Leah
Machine Learning
Identifying mistakes (i.e., miscues) made while reading aloud is commonly approached post-hoc by comparing automatic speech recognition (ASR) transcriptions to the target reading text. However, post-hoc methods perform poorly when ASR inaccurately transcribes verbatim speech. To improve on current methods for reading error annotation, we propose a novel end-to-end architecture that incorporates the target reading text via prompting and is trained for both improved verbatim transcription and direct miscue detection. Our contributions include: first, demonstrating that incorporating reading text through prompting benefits verbatim transcription performance over fine-tuning, and second, showing that it is feasible to augment speech recognition tasks for end-to-end miscue detection. We conducted two case studies -- children's read-aloud and adult atypical speech -- and found that our proposed strategies improve verbatim transcription and miscue detection compared to current state-of-the-art.
title Prompting Whisper for Improved Verbatim Transcription and End-to-end Miscue Detection
topic Machine Learning
url https://arxiv.org/abs/2505.23627