Matching Inconsistently Spelled Names in Automatic Speech Recognizer Output for Information Retrieval
Lists
Only you see this
Abstract
Many proper names are spelled inconsistently in speech recognizer output, posing a problem for applications where locating mentions of named entities is critical. We model the distortion in the spelling of a name due to the speech recognizer as the effect of a noisy channel. The models follow the framework of the IBM translation models. The model is trained using a parallel text of closed caption and automatic speech recognition output. We also test a string edit distance based method. The effectiveness of these models is evaluated on a name query retrieval task. Our methods result in a 60% improvement in F1. We also demonstrate why the problem has not been critical in TREC and TDT tasks.
BibTeX
@inproceedings{raghavan2005matching,
author = {Hema Raghavan and James Allan},
title = {{Matching Inconsistently Spelled Names in Automatic Speech Recognizer Output for Information Retrieval}},
booktitle = {Conference on Empirical Methods in Natural Language Processing},
year = {2005},
}| Generative Annotation for ASR Named Entity Correction | 2025 |
| Information Retrieval on Noisy Text | Infoscience · 2003 |
| An Approach to Using the Web as a Live Corpus for Spoken Transliteration Name Access | Int. J. Comput. Linguistics Chin. Lang. Process · 2006 |
| Spell My Name: Keyword Boosted Speech Recognition | arXiv · 2021 |
| DeRAGEC: Denoising Named Entity Candidates with Synthetic Rationale for ASR Error Correction | arXiv · 2025 |
| Robust Question Answering for Speech Transcripts Using Minimal Syntactic Analysis | Clef · 2007 |