Searcharxiv⌕ Search

arXiv subjects

Caterina Mauri

Publications and source records attributed to Caterina Mauri.

5 recordsLinked to original sources
arXiv2026

Coconstructions in spoken data: UD annotation guidelines and first results

The paper proposes annotation guidelines for syntactic dependencies that span across speaker turns - including collaborative coconstructions proper, wh-question answers, and backchannels - in spoken language treebanks within the Universal Dependencies framework. Two representations are proposed: a speaker-based representation following the segmentation into speech turns, and a dependency-based representation with dependencies across speech turns. New propositions are also put forward to distinguish between reformulations and repairs, and to promote elements in unfinished phrases.

cs.CL↗
arXiv2026

Is Semi-Automatic Transcription Useful in Corpus Creation? Preliminary Considerations on the KIParla Corpus

This paper analyses the implementation of Automatic Speech Recognition (ASR) into the transcription workflow of the KIParla corpus, a resource of spoken Italian. Through a two-phase experiment, 11 expert and novice transcribers produced both manual and ASR-assisted transcriptions of identical audio segments across three different types of conversation, which were subsequently analyzed through a combination of statistical modeling, word-level alignment and a series of annotation-based metrics. Results show that ASR-assisted workflows can increase transcription speed but do not consistently improve overall accuracy, with effects depending on multiple factors such as workflow configuration, conversation type and annotator experience. Analyses combining alignment-based metrics, descriptive statistics and statistical modeling provide a systematic framework to monitor transcription behavior across annotators and workflows. Despite limitations, ASR-assisted transcription, potentially supported by task-specific fine-tuning, could be integrated into the KIParla transcription workflow to accelerate corpus creation without compromising transcription quality.

cs.CL↗
arXiv2024

Did somebody say "Gest-IT"? A pilot exploration of multimodal data management

The paper presents a pilot exploration of the construction, management and analysis of a multimodal corpus. Through a three-layer annotation that provides orthographic, prosodic, and gestural transcriptions, the Gest-IT resource allows to investigate the variation of gesture-making patterns in conversations between sighted people and people with visual impairment. After discussing the transcription methods and technical procedures employed in our study, we propose a unified CoNLL-U corpus and indicate our future steps

cs.CL↗
arXiv2024

Beyond the Headlines: Understanding Sentiments and Morals Impacting Female Employment in Spain

After decades of improvements in the employment conditions of females in Spain, this process came to a sudden stop with the Great Spanish Recession of 2008. In this contribution, we analyse a large longitudinal corpus of national and regional news outlets employing advanced Natural Language Processing techniques to capture the valence of mentions of gender inequality expressed in the Spanish press. The automatic analysis of the news articles does indeed capture the known hardships faced by females in the Spanish labour market. Our approach can be straightforwardly generalised to other topics of interest. Assessing the sentiment and moral values expressed in the articles, we notice that females are, in the majority of cases, concerned more than males when there is a deterioration in the overall labour market conditions, based on newspaper articles. This behaviour has been present in the entire period of study (2000--2022) and looked particularly pronounced during the economic crisis of 2008 and the recent COVID-19 pandemic. Most of the time, this phenomenon looks to be more pronounced at the regional level, perhaps caused by a significant focus on local labour markets rather than on aggregate statistics or because, in local contexts, females might suffer more from an isolation or discrimination condition. Our findings contribute to a deeper understanding of the gender inequalities in Spain using alternative data, informing policymakers and stakeholders.

cs.CY↗