TEMPOS · Temporal Precision in Speech––Unveiling the Physiology of Prosodic Structure
„Хоризонт Европа“ — Действия „Мария Склодовска-Кюри“
- Период
- 2026-07-01 → 2028-06-30
- Финансиране от ЕС
- 217 965 €
- Участници
- 2
- Схема
- HORIZON-TMA-MSCA-PF-EF
Линиите свързват координатора с партньорите.
Накратко на български
Ритъмът и мелодията на речта се изследват чрез връзката между дишането и движенията на езика и устните. Това помага за подобряване на технологиите за разпознаване на глас, преподаването на чужди езици и клиничната диагностика.
Кратко обяснение, генерирано от езиков модел по текста на CORDIS. Оригиналът е по-долу.
Цел на проекта
Prosody – the rhythm and melody of speech – is fundamental to communication, shaping phrasing, emphasis, and meaning. Yet the physiological basis of how breathing and articulatory movements interact to shape prosodic rhythm has received little systematic attention. This lack of grounding also limits speech technologies such as automatic speech recognition (ASR), in handling variation across dialects, emotions, and second-language (L2) speech.TEMPOS (Temporal Precision in Speech – Unveiling the Physiology of Prosodic Structure) will address this by developing a new framework based on articulatory acceleration. In my previous work, I demonstrated that acceleration peaks – moments when force is added – provide stable, measurable landmarks of prosodic structure. Building on this, TEMPOS will combine electromagnetic articulography (EMA), respiratory inductance plethysmography (RIP), and perturbation experiments to examine how breathing cycles, articulatory stability, and prosodic rhythm interact in French, German, Swedish, and L2 speech.The project pursues three goals: (1) determine how breathing and syllable prominence co-depend across exhalation; (2) test posture stability and rhythm control under perturbation; and (3) build acceleration-based models linking articulation and acoustics for explainable AI. Outcomes will provide the first physiologically grounded account of prosodic rhythm, informing clinical diagnosis, L2 teaching, and speech technology.The project will be hosted at Humboldt-Universität zu Berlin (16 months) and GIPSA-lab, Grenoble (8 months), under the supervision of leading experts in phonetics and speech motor control, TEMPOS will advance methodological innovation and provide the pilot results needed for future competitive funding at national and international level.
Оригинален текст от CORDIS (на английски).
Участници
Връзки
Данни: CORDIS, © Европейски съюз