We describe ongoing work for developing a workflow for the applied use case of classifying diachronic and regional language variation in Pre-Modern Slavic texts. The data were obtained via handwritten text recognition (HTR) on medieval manuscripts and printings and partly by manual transcription. Our goal is to develop a workflow for such historical language data, covering HTR-postprocessing, annotating and classifying the digitized texts. We test and adapt existing language resources to fit the pipeline with low-barrier tooling, accessible for Humanists with limited experience in research data infrastructures, computational analysis or advanced methods of natural language processing (NLP). The workflow starts by addressing ground truth (GT) data creation for diagnosing and correcting HTR errors via string metrics and data-driven methods. On GT and on HTR data, we subsequently show classification results using transfer learning on sentence-level text snippets. Next, we report on our token-level data labeling efforts. Each step of the workflow is complemented with describing current limitations and our corresponding work in progress.

Our study presents a stratified dataset compiled from six different Slavic bodies of text, for cross-linguistic and diachronic analyses of Slavic Pre-Modern language variants. We demonstrate unsupervised domain adaptation and supervised finetuning of BERT on these low-resource, historical Slavic variants, for the purposes of provenance attribution in terms of three downstream tasks: manuscript, century and copying region classification.The data compilation aims to capture diachronic as well as regional language variation and change: the texts were written in the course of roughly a millennium, incorporating language variants from the High Middle Ages to the Early Modern Period, and originate from a variety of geographic regions. Mechanisms of language change in relatively small portions of such data have been inspected, analyzed and typologized by Slavists manually; our contribution aims to investigate the extent to which the BERT transformer architecture and pretrained models can benefit this process. Using these datasets for domain adaptation, we could attribute temporal, geographical and manuscript origin on the level of text snippets with high F-scores. We also conducted a qualitative analysis of the models’ misclassifications.

Since several decades emotional databases have been recorded by various laboratories. Many of them contain acted portrays of Darwin’s famous “big four” basic emotions. In this paper, we investigate in how far a selection of them are comparable by two approaches: on the one hand modeling similarity as performance in cross database machine learning experiments and on the other by analyzing a manually picked set of four acoustic features that represent different phonetic areas. It is interesting to see in how far specific databases (we added a synthetic one) perform well as a training set for others while some do not. Generally speaking, we found indications for both similarity as well as specificiality across languages.

To detect how and when readers are experiencing engagement with a literary work, we bring together empirical literary studies and language technology via focusing on the affective state of absorption. The goal of our resource development is to enable the detection of different levels of reading absorption in millions of user-generated reviews hosted on social reading platforms. We present a corpus of social book reviews in English that we annotated with reading absorption categories. Based on these data, we performed supervised, sentence level, binary classification of the explicit presence vs. absence of the mental state of absorption. We compared the performances of classical machine learners where features comprised sentence representations obtained from a pretrained embedding model (Universal Sentence Encoder) vs. neural classifiers in which sentence embedding vector representations are adapted or fine-tuned while training for the absorption recognition task. We discuss the challenges in creating the labeled data as well as the possibilities for releasing a benchmark corpus.

Current state-of-the-art speech synthesizers for domain-independent systems still struggle with the challenge of generating understandable and natural-sounding speech. This is mainly because the pronunciation of words of foreign origin, inflections and compound words often cannot be handled by rules. Furthermore there are too many of these for inclusion in exception dictionaries. We describe an approach to evaluating text-to-speech synthesizers with a subjective listening experiment. The focus is to differentiate between known problem classes for speech synthesizers. The target language is German but we believe that many of the described phenomena are not language specific. We distinguish the following problem categories: Normalization, Foreign linguistics, Natural writing, Language specific and General. Each of them is divided into five to three problem classes. Word lists for each of the above mentioned categories were compiled and synthesized by both a commercial and an open source synthesizer, both being based on the non-uniform unit-selection approach. The synthesized speech was evaluated by human judges using the Speechalyzer toolkit and the results are discussed. It shows that, as expected, the commercial synthesizer performs much better than the open-source one, and especially words of foreign origin were pronounced badly by both systems.

The BAS CLARIN speech data repository is introduced. At the current state it comprises 31 pre-dominantly German corpora of spoken language. It is compliant to the CLARIN-D as well as the OLAC requirements. This enables its embedding into several infrastructures. We give an overview over its structure, its implementation as well as the corpora it contains.

In 2012 the Bavarian Archive for Speech Signals started providing some of its tools from the field of spoken language in the form of Software as a Service (SaaS). This means users access the processing functionality over a web browser and therefore do not have to install complex software packages on a local computer. Amongst others, these tools include segmentation & labeling, grapheme-to-phoneme conversion, text alignment, syllabification and metadata generation, where all but the last are available for a variety of languages. Since its creation the number of available services and the web interface have changed considerably. We give an overview and a detailed description of the system architecture, the available web services and their functionality. Furthermore, we show how the number of files processed over the system developed in the last four years.

We present a data-driven method for determining the veracity of a set of rumorous claims on social media data. Tweets from different sources pertaining to a rumor are processed on three levels: first, factuality values are assigned to each tweet based on four textual cue categories relevant for our journalism use case; these amalgamate speaker support in terms of polarity and commitment in terms of certainty and speculation. Next, the proportions of these lexical cues are utilized as predictors for tweet certainty in a generalized linear regression model. Subsequently, lexical cue proportions, predicted certainty, as well as their time course characteristics are used to compute veracity for each rumor in terms of the identity of the rumor-resolving tweet and its binary resolution value judgment. The system operates without access to extralinguistic resources. Evaluated on the data portion for which hand-labeled examples were available, it achieves .74 F1-score on identifying rumor resolving tweets and .76 F1-score on predicting if a rumor is resolved as true or false.

The utilization of social media material in journalistic workflows is increasing, demanding automated methods for the identification of mis- and disinformation. Since textual contradiction across social media posts can be a signal of rumorousness, we seek to model how claims in Twitter posts are being textually contradicted. We identify two different contexts in which contradiction emerges: its broader form can be observed across independently posted tweets and its more specific form in threaded conversations. We define how the two scenarios differ in terms of central elements of argumentation: claims and conversation structure. We design and evaluate models for the two scenarios uniformly as 3-way Recognizing Textual Entailment tasks in order to represent claims and conversation structure implicitly in a generic inference model, while previous studies used explicit or no representation of these properties. To address noisy text, our classifiers use simple similarity features derived from the string and part-of-speech level. Corpus statistics reveal distribution differences for these features in contradictory as opposed to non-contradictory tweet relations, and the classifiers yield state of the art performance.