International audience

While large multilingual automatic speech recognition (ASR) models achieve remarkable performance, the internal mechanisms of the end-to-end pipeline, particularly concerning fairness and efficacy across languages, remain underexplored.This paper introduces a finegrained analysis of Whisper's multilingual decoder, examining its sub-token hypotheses during transcription across languages with various resource levels.Our method traces the beam search path, capturing sub-token guesses and their associated probabilities.Results reveal that higher resource languages benefit from higher likelihood of the correct token being topranked, greater confidence, lower predictive entropy, and more diverse alternative candidates.Lower resource languages fare worse on these metrics, but also exhibit distinct clustering patterns in sub-token usage sometimes influenced by typology in our PCA and t-SNE analysis.This sub-token probing uncovers systematic decoding disparities masked by aggregate error rates and points towards targeted interventions to ameliorate the imbalanced development of speech technology.

Essay writing is a skill commonly taught and practised in schools. The ability to write a fluent and persuasive essay is often a major component of formal assessment. In natural language processing and education technology we may work with essays in their final form, for example to carry out automated assessment or grammatical error correction. In this work we collect and analyse data representing the essay writing process from start to finish, by recording every key stroke from multiple writers participating in our study. We describe our data collection methodology, the characteristics of the resulting dataset, and the assignment of proficiency levels to the texts. We discuss the ways the keystroke data can be used – for instance seeking to identify patterns in the keystrokes which might act as features in automated assessment or may enable further advancements in writing assistance – and the writing support technology which could be built with such information, if we can detect when writers are struggling to compose a section of their essay and offer appropriate intervention. We frame this work in the context of English language learning, but we note that keystroke logging is relevant more broadly to text authoring scenarios as well as cognitive or linguistic analyses of the writing process.

This paper describes MAKE-NMTViz, a project designed to help translators visualize neural machine translation outputs using explainable artificial intelligence visualization tools initially developed for computer vision.

This paper describes work in progress on Visualisation tools to foster collaborations between translators and computational scientists. We aim to describe how visualisation features can be used to explain translation and NMT outputs. We tested several visualisation functionalities with three NMT models based on Chinese-English, Spanish-English and French-English language pairs. We created three demos containing different visualisation tools and analysed them within the framework of performance-explainability, focusing on the translator’s perspective.

International audience

International audience

In this paper, we reproduce some of the experiments related to neural network training for Machine Translation as reported in (Vanmassenhove and Way, 2018). They annotated a sample from the EN-FR and EN-DE Europarl aligned corpora with syntactic and semantic annotations to train neural networks with the Nematus Neural Machine Translation (NMT) toolkit. Following the original publication, we obtained lower BLEU scores than the authors of the original paper, but on a more limited set of annotations. In the second half of the paper, we try to analyze the difference in the results obtained and suggest some methods to improve the results. We discuss the Byte Pair Encoding (BPE) used in the pre-processing phase and suggest feature ablation in relation to the granularity of syntactic and semantic annotations. The learnability of the annotated input is discussed in relation to existing resources for the target languages. We also discuss the feature representation likely to have been adopted for combining features.

This paper presents an annotation framework for nasal grunts of the whole French CID corpus (Bertrand et al., 2008). The acoustic components under scrutiny are justified and the annotation guidelines are described. We carefully characterise the acoustic cues and visual cues followed by the annotator, especially for non-modal phonation types. The conventions followed for the annotation of interactional and positional properties of grunts are explained. The resulting datasets after data extraction with Praat scripts (Boersma and Weenink, 2019) are analysed with R (R Core Team, 2017), focusing on duration. We analyse the effect of non-modal phonation (especially ingressive phonation) on duration and discuss a specialisation of grunts observed in the CID for grunts with ingressive phonation. The more general aim of this research is to establish putative core and additive properties of grunts and a tentative typology of grunts in spoken interactions.