Publication detail

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

BENEŠ, K. KOCOUR, M. BURGET, L.

Original Title

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

Type

conference paper

Language

English

Original Abstract

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.

Keywords

confidences measures, system fusion, end-toend systems, automatic speech recognition

Authors

BENEŠ, K.; KOCOUR, M.; BURGET, L.

Released

14. 4. 2024

Publisher

IEEE Signal Processing Society

Location

Seoul

ISBN

979-8-3503-4485-1

Book

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

Pages from

11276

Pages to

11280

Pages count

5

URL

https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10446739

BibTex

@inproceedings{BUT189696,
  author="Karel {Beneš} and Martin {Kocour} and Lukáš {Burget}",
  title="Hystoc: Obtaining Word Confidences for Fusion of End-To-End ASR Systems",
  booktitle="ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
  year="2024",
  pages="11276--11280",
  publisher="IEEE Signal Processing Society",
  address="Seoul",
  doi="10.1109/ICASSP48485.2024.10446739",
  isbn="979-8-3503-4485-1",
  url="https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10446739"
}