Some of my projects

More information about my projects are available on the site of the CAI - Center for Augmented Interpretation of the university of Mainz.

Evaluating Automatic Spoken Language Translation
Design of a user-centric linguistic framework to evaluate speech-to-text and speech-to-speech translation from a communicative perspective and empirical evaluating human and machine interpreting.

Webpage: M.INTerpreting
Publication: Fantinuoli C., Prandi B. "Towards the evaluation of automatic simultaneous speech translation from a communicative perspective". Proceedings of IWSLT 2021 (2021)
Grant: Inneruniversitäre Forschungsforderung 2020-2021
ABM-SCI: AI-tool for teching and research
The first Artificial Boothmate designed for use in the Interpreting Classroom and for Research purposes. It suggests interpreters in real-time terms, translations, and numbers in context.

Project's webpage: info
EasyAI: Explaining AI and NLP to the humanities
A minimalist course on Natural Language Processing and Artificial Intelligence designed for Translators and Interpreters.

Project's webpage: here
Grant: GLK Mainz
Monitoring the interpretability of speech
Monitoring the interpretability of speech
Real-time, algorithmic monitoring of interpretability of a speech and visual signalling system for the speaker.

Project's webpage: Speedometer
Monitoring system: prototype
Patent pending - KUDO/Fantinuoli
Neural system to improve English text
This system uses a language models to reformulate sentences, correcting errors such as grammar, collocations, etc.

Project's webpage: demo
ASR Computer assisted interpreting
Augmented Interpretation
AI-enanced computer-assisted interpreting tools and the empirical study of the human-machine interaction from an interpreting studies perspective.

Project's webpage: Computer-assisted interpreting tools
Demo of ASR-supported CAI tool for numbers: prototype
automatic speech translation
Automatic Speech-to-Text Translation
We have developed a baseline speech translation engine for real-time continuous (simultaneous) modality. The goal is to develope a baseline engine to perform tests on state-of-the-art speech translation and benchmark other commercial and academic engines.

Live demo: Automatic Speech-to-Text Translation
Ergonomics of the artificial boothmate
With the University Gent, we are working on understanding the human-machine interaction of ASR-based computer-assisted tools and measuring its use on interpreting performance.

Project's webpage: EABM project
Video Intro: here
Grant: N° EC 10 – 2020-2021 – SCIC.B.1(2020) 4773772 - European Commission
GEROM: A German-Italian, corpus-based terminology platform for social discourse analysis
Grant: DAAD