A5076040879- Natural Language Processing Techniques
- Speech Recognition and Synthesis
- Topic Modeling
- Linguistics and Discourse Analysis
- French Language Learning Methods
- Text Readability and Simplification
- Speech and dialogue systems
- Safety Warnings and Signage
- Tactile and Sensory Interactions
- Handwritten Text Recognition Techniques
- Subtitles and Audiovisual Media
Laboratoire d'Informatique de Grenoble
2022-2024
Grenoble Images Parole Signal Automatique
2022-2024
Université Grenoble Alpes
2022-2023
Research efforts in syntactic parsing have focused on written texts.As a result, speech is usually performed transcriptions, either unrealistic settings (gold transcriptions) or predicted transcriptions.Parsing from though straightforward to implement using out-of-the-box tools for Automatic Speech Recognition (ASR) and dependency has two important limitations.First, relying transcriptions will lead error propagation due recognition mistakes.Secondly, many acoustic cues that are (prosody,...
Direct dependency parsing of the speech signal -- as opposed to transcriptions has recently been proposed a task (Pupier et al. 2022), way incorporating prosodic information in system and bypassing limitations pipeline approach that would consist using first an Automatic Speech Recognition (ASR) then syntactic parser. In this article, we report on set experiments aiming at assessing performance two paradigms (graph-based sequence labeling based parsing) parsing. We perform evaluation large...
Self-supervised learning (SSL) is at the origin of unprecedented improvements in many different domains including computer vision and natural language processing. Speech processing drastically benefitted from SSL as most current domain-related tasks are now being approached with pre-trained models. This work introduces LeBenchmark 2.0 an open-source framework for assessing building SSL-equipped French speech technologies. It includes documented, large-scale heterogeneous corpora up to 14,000...