Publication Details
Written Term Detection Improves Spoken Term Detection
SARAÇLAR, M.
Keyword search, spoken term detection, keyword spotting, end-to-end keyword
search, multitask learning, domain adaptation, masked language modeling.
End-to-end (E2E) approaches to keyword search (KWS) are considerably simpler in
terms of training and indexing complexity when compared to approaches which use
the output of automatic speech recognition (ASR) systems. This simplification
however has drawbacks due to the loss of modularity. In partic- ular, where
ASR-based KWS systems can benefit from external unpaired text via a language
model, current formulations of E2E KWS systems have no such mechanism. Therefore,
in this paper, we propose a multitask training objective which allows unpaired
text to be integrated into E2E KWS without complicating indexing and search. In
addition to training an E2E KWS model to retrieve text queries from spoken
documents, we jointly train it to retrieve text queries from masked written
documents. We show empirically that this approach can effectively leverage
unpaired text for KWS, with significant improvements in search performance across
a wide variety of languages. We conduct analysis which indicates that these
improvements are achieved because the proposed method improves document
representations for words in the unpaired text. Finally, we show that the
proposed method can be used for domain adaptation in settings where in-domain
paired data is scarce or nonexistent.
@article{BUT193391,
author="YUSUF, B. and SARAÇLAR, M.",
title="Written Term Detection Improves Spoken Term Detection",
journal="IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH AND LANGUAGE PROCESSING",
year="2024",
volume="32",
number="06",
pages="3213--3223",
doi="10.1109/TASLP.2024.3407476",
issn="2329-9290",
url="https://ieeexplore.ieee.org/document/10571348"
}