Hystoc: Obtaining Word Confidences for Fusion of End-To-End ASR Systems

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F24%3APU152209" target="_blank" >RIV/00216305:26230/24:PU152209 - isvavai.cz</a>

Výsledek na webu

<a href="https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10446739" target="_blank" >https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10446739</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICASSP48485.2024.10446739" target="_blank" >10.1109/ICASSP48485.2024.10446739</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Hystoc: Obtaining Word Confidences for Fusion of End-To-End ASR Systems

Popis výsledku v původním jazyce

End-to-end (e2e) systems have recently gained wide popularity in automatic speech recognition. However, these systems do generally not provide well-calibrated word-level confidences. In this paper, we propose Hystoc, a simple method for obtaining word-level confidences from hypothesis-level scores. Hystoc is an iterative alignment procedure which turns hypotheses from an n-best output of the ASR system into a confusion network. Eventually, word-level confidences are obtained as posterior probabilities in the individual bins of the confusion network. We show that Hystoc provides confidences that correlate well with the accuracy of the ASR hypothesis. Furthermore, we show that utilizing Hystoc in fusion of multiple e2e ASR systems increases the gains from the fusion by up to 1% WER absolute on Spanish RTVE2020 dataset. Finally, we experiment with using Hystoc for direct fusion of n-best outputs from multiple systems, but we only achieve minor gains when fusing very similar systems.

Název v anglickém jazyce

Hystoc: Obtaining Word Confidences for Fusion of End-To-End ASR Systems

Popis výsledku anglicky

End-to-end (e2e) systems have recently gained wide popularity in automatic speech recognition. However, these systems do generally not provide well-calibrated word-level confidences. In this paper, we propose Hystoc, a simple method for obtaining word-level confidences from hypothesis-level scores. Hystoc is an iterative alignment procedure which turns hypotheses from an n-best output of the ASR system into a confusion network. Eventually, word-level confidences are obtained as posterior probabilities in the individual bins of the confusion network. We show that Hystoc provides confidences that correlate well with the accuracy of the ASR hypothesis. Furthermore, we show that utilizing Hystoc in fusion of multiple e2e ASR systems increases the gains from the fusion by up to 1% WER absolute on Spanish RTVE2020 dataset. Finally, we experiment with using Hystoc for direct fusion of n-best outputs from multiple systems, but we only achieve minor gains when fusing very similar systems.

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

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

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

11276-11280

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

—