Background: The preservation and analysis of spoken data in oral history archives, such as Holocaust testimonies, provide a vast and complex knowledge source. These archives pose unique challenges and opportunities for computational methods, particularly in self-supervised learning and information retrieval.Objective: This study explores the application of state-of-the-art artificial intelligence (AI) models, particularly transformer-based architectures, to enhance navigation and engagement with large-scale oral history testimonies. The goal is to improve accessibility while preserving the authenticity and integrity of historical records.Methods: We developed an asking questions framework utilizing a fine-tuned T5 model to generate contextually relevant questions from interview transcripts. To ensure semantic coherence, we introduced a semantic continuity model based on a BERT-like architecture trained with contrastive loss.Results: The system successfully generated contextually relevant questions from oral history testimonies, enhancing user navigation and engagement. Filtering techniques improved question quality by retaining only semantically coherent outputs, ensuring alignment with the testimony content. The approach demonstrated effectiveness in handling spontaneous, unstructured speech, with a significant improvement in question relevance compared to models trained on structured text. Applied to real-world interview transcripts, the framework balanced enrichment of user experience with preservation of historical authenticity.Conclusion: By integrating generative AI models with robust retrieval techniques, we enhance the accessibility of oral history archives while maintaining their historical integrity. This research demonstrates how AI-driven approaches can facilitate interactive exploration of vast spoken data repositories, benefiting researchers, historians and the general public.

In this paper, we are comparing monolingual Wav2Vec 2.0 models with various multilingual models to see whether we could improve speech recognition performance on a unique oral history archive containing a lot of mixed-language sentences. Our main goal is to push forward research on this unique dataset, which is an extremely valuable part of our cultural heritage. Our results suggest that monolingual speech recognition models are, in most cases, superior to multilingual models, even when processing the oral history archive full of mixed-language sentences from non-native speakers. We also performed the same experiments on the public CommonVoice dataset to verify our results. We are contributing to the research community by releasing our pre-trained models to the public.

Oral history is about oral sources of witnesses and commentors on historical events. Speech technology is an important instrument to process such recordings in order to obtain transcription and further enhancements to structure the oral account In this contribution we address the transcription portal and the webservices associated with speech processing at BAS, speech solutions developed at LINDAT, how to do it yourself with Whisper, remaining challenges, and future developments.

The paper introduces software capable of indexing and searching large archives of scanned historical documents. The system capabilities are demonstrated on the collection containing documents from the archives of the post-Soviet security services. The backend of the system was designed with a focus on flexibility (it is actually already being used for other related tasks) and scalability to larger volumes of data. The graphical user interface design has been consulted with historians interested in using the archived documents and was developed in several iterations, gradually including the changes induced both by the user’s requests and by our improving knowledge about the nature of the processed data.

This paper is a step forward in our effort to make vast oral history archives more accessible to the public and researchers by breaking down the decoding barriers between the knowledge encoded in the spoken testimonies and users who want to search for the information of their interest. We present new Transformer-based monolingual models suitable for speech recognition of oral history archives in English, German, and Czech. Our experiments show that although the all-purpose speech recognition systems have recently made tremendous progress, the transcription of oral history archives is still a challenging task for them; our tailored models significantly outper-formed larger public multilingual models and scored new state-of-the-art results on all tested datasets. Due to the 2-phase fine-tuning process, our models are robust and can be used for oral history archives of various domains. We publicly release our models within a public speech recognition service.

This paper discusses the analogies between the mainstream theory of human mind and the two broad paradigms that are employed when building artificial intelligence systems. Then it ponders the idea how those analogies could be utilized in building a truly explainable artificial intelligence (AI) applications. The core part is devoted to the problem of unwanted rationalization that could disguise the true reasons lying behind the decisions of the explainable AI systems.

This work presents a pipeline for processing digitally scanned documents, reading their textual content, and storing it in a dataset for the purpose of information retrieval. The pipeline is able to handle images of various quality, whether they were obtained by a digital scanner or camera. The image can contain multiple pages in any layout, but an approximate upright orientation is assumed. The pipeline uses Faster R-CNN to detect individual pages. These are then processed by a deskew algorithm to correct the orientation, and finally read by the Tesseract OCR system that has been retrained on a large set of synthetic images and a small set of annotated real-world documents. By applying the pipeline, we were able to increase the word recall to 60.56% which is an absolute gain of 19.19% from the baseline solution that uses only Tesseract OCR. A demo of the proposed pipeline can be found at https://archivkgb.zcu.cz/.

The paper proposes a module for automatic punctuation prediction and casing reconstruction based on transformers architectures (BERT/T5) that constitutes the current state-of-the-art in many similar NLP tasks. The main motivation for our work was to increase the readability of the ASR output. The ASR output is usually in the form of a continuous stream of text, without punctuation marks and with all words in lowercase. The resulting punctuation and casing reconstruction module is evaluated on both the written text and the actual ASR output in three languages (English, Czech and Slovak).

In this paper we propose a pipeline for processing of scanned historical documents into the electronic text form that could then be indexed and stored in a database. The nature of the documents presents a substantial challenge for standard automated techniques – not only there is a mix of typewritten and handwritten documents of varying quality but the scanned pages often contain multiple documents at once. Moreover, the language of the texts alternates mostly between Russian and Ukrainian but other languages also occur. The paper focuses mainly on segmentation, document type classification, and image preprocessing of the scanned documents; the output of those methods is then passed to the off-the-shelf OCR software and a baseline performance is evaluated on a simplified OCR task.

A richly annotated and genre-diversified language resource, The Prague Dependency Treebank – Consolidated 1.0 (PDT-C 1.0, or PDT-C in short in the sequel) is a consolidated release of the existing PDT-corpora of Czech data, uniformly annotated using the standard PDT scheme. PDT-corpora included in PDT-C: Prague Dependency Treebank (the original PDT contents, written newspaper and journal texts from three genres); Czech part of Prague Czech-English Dependency Treebank (translated financial texts, from English), Prague Dependency Treebank of Spoken Czech (spoken data, including audio and transcripts and multiple speech reconstruction annotation); PDT-Faust (user-generated texts). The difference from the separately published original treebanks can be briefly described as follows: it is published in one package, to allow easier data handling for all the datasets; the data is enhanced with a manual linguistic annotation at the morphological layer and new version of morphological dictionary is enclosed; a common valency lexicon for all four original parts is enclosed. Documentation provides two browsing and editing desktop tools (TrEd and MEd) and the corpus is also available online for searching using PML-TQ.

In this paper, we present our experiments with BERT (Bidirectional Encoder Representations from Transformers) models in the task of sentiment analysis, which aims to predict the sentiment polarity for the given text. We trained an ensemble of BERT models from a large self-collected movie reviews dataset and distilled the knowledge into a single production model. Moreover, we proposed an improved BERT's pooling layer architecture, which outperforms standard classification layer while enables per-token sentiment predictions. We demonstrate our improvements on a publicly available dataset with Czech movie reviews.

The paper introduces the motivation for creating dedicated speech corpora of air traffic control communication, describes in detail the process of preparation of corpora for both automatic speech recognition and text-to-speech synthesis, presents an illustrative example of speech recognition system developed using the automatic speech recognition corpora and finally describes the technical aspects of the data and the distribution channel.

The aim of this article is to show that the document embedding using the doc2vec algorithm can substantially improve the performance of the standard method for unsupervised document classification – the K-means clustering. We have performed rather extensive set of experiments on one English and two Czech datasets and the results suggest that representing the documents using vectors generated by the doc2vec algorithm brings a consistent improvement across languages and datasets. The English dataset – 20NewsGroups – was processed in a way that allows direct comparison with the results of both supervised and unsupervised algorithms published previously. Such comparison is provided in the paper, together with the results of supervised classification achieved by the state-of-the-art SVM classifier.

This paper explores the possibility to use grapheme-based word and sub-word models in the task of spoken term detection (STD). The usage of grapheme models eliminates the need for expert-prepared pronunciation lexicons (which are often far from complete) and/or trainable grapheme-to-phoneme (G2P) algorithms that are frequently rather inaccurate, especially for rare words (words coming from a different language). Moreover, the G2P conversion of the search terms that need to be performed on-line can substantially increase the response time of the STD system. Our results show that using various grapheme-based models, we can achieve STD performance (measured in terms of ATWV) comparable with phoneme-based models but without the additional burden of G2P conversion.

In this paper, we describe the initial stages of our project, the goal of which is to create an integrated archive of the recordings, scanned documents, and photographs that would be accessible online and would provide multifaceted search capabilities (spoken content, biographical information, relevant time period, etc.). The recordings contain retrospective interviews with the witnesses of the totalitarian regimes in Czechoslovakia, where the vocabulary used in such interviews consists of many archaic words and named entities that are now quite rare in everyday speech. The scanned documents consist of text materials and photographs mainly from the home archives of the interviewees or the archive of the State Security. These documents are usually typewritten or even handwritten and have really bad optical quality. In order to build an integrated archive, we will employ mainly methods of automatic speech recognition (ASR), automatic indexing and search in recognized recordings and, to a certain extent, also the optical character recognition (OCR). Other natural language processing techniques like topic detection are also planned to be used in the later stages of the project. This paper focuses on the processing of the speech data using ASR and the scanned typewritten documents with OCR and describes the initial experiments.

The paper describes the process of creation of domain-specific speech corpora containing air traffic control (ATC) communication prompts. Since the ATC domain is highly specific both from the acoustic point-of-view (significant level of noise in the signal, non-native English accents of the speakers, non-standard pronunciation of some frequent words) and the lexical and syntactic perspective (prescribed structure of utterances, rather limited vocabulary), it is useful to collect and annotate data from this specific domain. Actually, the ultimate goal of the research effort of our team was to develop a voice dialogue system simulating the responses of the pilot that could be used for training aspiring air traffic controllers. In order to do so, we needed - among other modules - a domain-specific automatic speech recognition (ASR) and text-to-speech synthesis (TTS) engines. This paper concentrates on the details of the ASR and TTS corpora creation process but also overviews their usage in preparing practical applications and provides links to the distribution channel of the data.

This article presents a method for pre-processing the feature vectors representing text documents that are consequently classified using unsupervised methods. The main goal is to show that state-of-the-art classification methods can be improved by a certain data preparation process. The first method is a standard K-means clustering and the second Latent Dirichlet allocation (LDA) method. Both are widely used in text processing. The mentioned algorithms are applied to two data sets in two different languages. First of them, the 20NewsGroup is a widely used benchmark for classification of English documents. The second set was selected from the large body of Czech news articles and was used mainly to compare the performance of the tested methods also for the case of less frequently studied language. Furthermore, the unsupervised methods are also compared with the supervised ones in order to (in some sense) ascertain the upper-bound of the task.

We summarize the involvement of our CEMI team in the ”NLI Shared Task 2017”, which deals with both textual and speech input data. We submitted the results achieved by using three different system architectures; each of them combines multiple supervised learning models trained on various feature sets. As expected, better results are achieved with the systems that use both the textual data and the spoken responses. Combining the input data of two different modalities led to a rather dramatic improvement in classification performance. Our best performing method is based on a set of feed-forward neural networks whose hidden-layer outputs are combined together using a softmax layer. We achieved a macro-averaged F1 score of 0.9257 on the evaluation (unseen) test set and our team placed first in the main task together with other three teams.

PDTSC 1.0 is a multi-purpose corpus of spoken language. 768,888 tokens, 73,374 sentences and 7,324 minutes of spontaneous dialog speech have been recorded, transcribed and edited in several interlinked layers: audio recordings, automatic and manual transcription and manually reconstructed text. PDTSC 1.0 is a delayed release of data annotated in 2012. It is an update of Prague Dependency Treebank of Spoken Language (PDTSL) 0.5 (published in 2009). In 2017, Prague Dependency Treebank of Spoken Czech (PDTSC) 2.0 was published as an update of PDTSC 1.0.