Contextual Biasing Methods for Improving Rare Word Detection in Automatic Speech Recognition

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F24%3APU154700" target="_blank" >RIV/00216305:26230/24:PU154700 - isvavai.cz</a>

Výsledek na webu

<a href="https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10447465" target="_blank" >https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10447465</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICASSP48485.2024.10447465" target="_blank" >10.1109/ICASSP48485.2024.10447465</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Contextual Biasing Methods for Improving Rare Word Detection in Automatic Speech Recognition

Popis výsledku v původním jazyce

In specialized domains like Air Traffic Control (ATC), a notable challenge in porting a deployed Automatic Speech Recognition (ASR) system from one airport to another is the alteration in the set of crucial words that must be ac- curately detected in the new environment. Typically, such words have limited occurrences in training data, making it impractical to retrain the ASR system. This paper explores innovative word-boosting techniques to improve the detec- tion rate of such rare words in the ASR hypotheses for the ATC domain. Two acoustic models are investigated: a hybrid CNN-TDNNF model trained from scratch and a pre-trained wav2vec2-based XLSR model fine-tuned on a common ATC dataset. The word boosting is done in three ways. First, an out-of-vocabulary word addition method is explored. Second, G-boosting is explored, which amends the language model before building the decoding graph. Third, the boosting is performed on the fly during decoding using lattice re-scoring. The results indicate that the G-boosting method performs best and provides an approximately 30-43% relative improvement in recall of the boosted words. Moreover, a relative improve- ment of up to 48% is obtained upon combining G-boosting and lattice-rescoring

Název v anglickém jazyce

Contextual Biasing Methods for Improving Rare Word Detection in Automatic Speech Recognition

Popis výsledku anglicky

In specialized domains like Air Traffic Control (ATC), a notable challenge in porting a deployed Automatic Speech Recognition (ASR) system from one airport to another is the alteration in the set of crucial words that must be ac- curately detected in the new environment. Typically, such words have limited occurrences in training data, making it impractical to retrain the ASR system. This paper explores innovative word-boosting techniques to improve the detec- tion rate of such rare words in the ASR hypotheses for the ATC domain. Two acoustic models are investigated: a hybrid CNN-TDNNF model trained from scratch and a pre-trained wav2vec2-based XLSR model fine-tuned on a common ATC dataset. The word boosting is done in three ways. First, an out-of-vocabulary word addition method is explored. Second, G-boosting is explored, which amends the language model before building the decoding graph. Third, the boosting is performed on the fly during decoding using lattice re-scoring. The results indicate that the G-boosting method performs best and provides an approximately 30-43% relative improvement in recall of the boosted words. Moreover, a relative improve- ment of up to 48% is obtained upon combining G-boosting and lattice-rescoring

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ISBN

979-8-3503-4485-1

ISSN

—

e-ISSN

—

Počet stran výsledku

5

Strana od-do

12652-12656

Název nakladatele

IEEE Signal Processing Society

Místo vydání

Seoul

Místo konání akce

Seoul

Datum konání akce

14. 4. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—