Languages may be differently distant from each other and their mutual intelligibility may be asymmetric. In this paper we introduce incom.py, a toolbox for calculating linguistic distances and asymmetries between related languages. incom.py allows linguist experts to quickly and easily perform statistical analyses and compare those with experimental results. We demonstrate the efficacy of incom.py in an incomprehension experiment on two Slavic languages: Bulgarian and Russian. Using incom.py we were able to validate three methods to measure linguistic distances and asymmetries: Levenshtein distance, word adaptation surprisal, and conditional entropy as predictors of success in a reading intercomprehension experiment.

Focusing on orthography as a primary linguistic interface in every reading activity, the central research question we address here is how orthographic intelligibility can be measured and predicted between closely related languages. This paper presents methods and findings of modeling orthographic intelligibility in a reading intercomprehension scenario from the information-theoretic perspective. The focus of the study is on two Slavic language pairs: Czech–Polish (West Slavic, using the Latin script) and Bulgarian–Russian (South Slavic and East Slavic, respectively, using the Cyrillic script). In this article, we present computational methods for measuring orthographic distance and orthographic asymmetry by means of the Levenshtein algorithm, conditional entropy and adaptation surprisal method that are expected to predict the influence of orthography on mutual intelligibility in reading.

Within the first shared task on machine translation between similar languages, we present our first attempts on Czech to Polish machine translation from an intercomprehension perspective. We propose methods based on the mutual intelligibility of the two languages, taking advantage of their orthographic and phonological similarity, in the hope to improve over our baselines. The translation results are evaluated using BLEU. On this metric, none of our proposals could outperform the baselines on the final test set. The current setups are rather preliminary, and there are several potential improvements we can try in the future.

State-of-the-art spoken language identification (LID) systems, which are based on end-to-end deep neural networks, have shown remarkable success not only in discriminating between distant languages but also between closely-related languages or even different spoken varieties of the same language. However, it is still unclear to what extent neural LID models generalize to speech samples with different acoustic conditions due to domain shift. In this paper, we present a set of experiments to investigate the impact of domain mismatch on the performance of neural LID systems for a subset of six Slavic languages across two domains (read speech and radio broadcast) and examine two low-level signal descriptors (spectral and cepstral features) for this task. Our experiments show that (1) out-of-domain speech samples severely hinder the performance of neural LID models, and (2) while both spectral and cepstral features show comparable performance within-domain, spectral features show more robustness under domain mismatch. Moreover, we apply unsupervised domain adaptation to minimize the discrepancy between the two domains in our study. We achieve relative accuracy improvements that range from 9% to 77% depending on the diversity of acoustic conditions in the source domain.

Sharing portions of grammars across languages greatly reduces the costs of nutltilingual grammar engineering. Related languages share a ntuch wider range of linguistic itff'ornuttio;t than typically assunwd in stamlard mttltilingttal gramtmtr atwhitectures. Taking grammatical relatedness seriously, we are particularly interested in designing lhtguistically motivated grammatical resottrces Jbr Slavic languages to be used itz applied and theoretical computational linguistics, ht order to gain the pelwpective of a language-family oriented gramntar desigtl, we consider {ttt arrtly of systematic relations that can hold between syntactical units. While the categorisation of primitive linguistic entities tends to be language-specilic or even constrttction-.~pecific, the relations holding between them allow viii'lofts degrees of absltztction. On the basis of Slavic &tta, we show how a domain ontology conceptualising molphosyntaclic "buildiltg blocks" can serve as a basis r~" a shared grotlt;nar of Slavic.