15 citations · 18 across the 6 of their papers we have counts for
7 papers
TextBite: A Historical Czech Document Dataset for Logical Page Segmentation
Martin Kostelník, Karel Beneš, Michal Hradiš
Logical page segmentation is an important step in document analysis, enabling better semantic representations, information retrieval, and text understanding. Previous approaches de…
BUT CHiME-7 system description
Martin Karafiát, Karel Veselý, Igor Szöke +5
This paper describes the joint effort of Brno University of Technology (BUT), AGH University of Krakow and University of Buenos Aires on the development of Automatic Speech Recogni…
Hystoc: Obtaining word confidences for fusion of end-to-end ASR systems
Karel Beneš, Martin Kocour, Lukáš Burget
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 confidence…
SoftCTC -- Semi-Supervised Learning for Text Recognition using Soft Pseudo-Labels
Martin Kišš, Michal Hradiš, Karel Beneš +2
This paper explores semi-supervised training for sequence tasks, such as Optical Character Recognition or Automatic Speech Recognition. We propose a novel loss function $\unicode{x…
Importance of Textlines in Historical Document Classification
Martin Kišš, Jan Kohút, Karel Beneš +1
This paper describes a system prepared at Brno University of Technology for ICDAR 2021 Competition on Historical Document Classification, experiments leading to its design, and the…
AT-ST: Self-Training Adaptation Strategy for OCR in Domains with Limited Transcriptions
Martin Kišš, Karel Beneš, Michal Hradiš
This paper addresses text recognition for domains with limited manual annotations by a simple self-training strategy. Our approach should reduce human annotation effort when target…
Text Augmentation for Language Models in High Error Recognition Scenario
Karel Beneš, Lukáš Burget
We examine the effect of data augmentation for training of language models for speech recognition. We compare augmentation based on global error statistics with one based on per-wo…