Seminal
The magazine and atlas of EU-funded researchA new publication from the web and AI team at Scott Charnley

Atlas / Italy

Marie Skłodowska-Curie Action · Ended early

COMUTTI Engineering voice-based models and interfaces for enhancing the speech therapy of minimally-verbal children with autism and their communication

Coordinator
Politecnico Di Milano, Italy
Runs
1 September 2024 to 31 August 2026
Status
Ended early
EU funding
€176k
Total cost
—
Partners
2 organisations in 2 countries
Type
Marie Skłodowska-Curie Action
Grant agreement
101109600

What the project is doing

In the project's own words, from its summary on CORDIS.

About 30% of children with autism are minimally verbal (MV), meaning they can communicate mainly through nonverbal vocalizations (i.e., vocalizations that do not have typical verbal content). Vocalizations often have self-consistent phonetic content and vary in tone, pitch, and duration depending on the individual's emotional state or intended communication. While vocalizations contain important affective and communicative information and are comprehensible by close caregivers, they are often poorly understood by those who do not know the communicator well.

An improved understanding of nonverbal vocalizations could pave the way for a better understanding of the cognitive, social, and emotional mechanisms associated with MV children with autism. Moreover, it could lead to new therapeutic interventions for these subjects based on advanced voice-based technology for Augmented Alternative Communication (AAC), i.e., for communicating without using words. This MSCA touches on the research fields of audio signal processing (in particular, children's vocalizations perception) and human-computer interaction (in particular, voice-based interfaces for speech therapy).

It aims to advance the understanding of MV autistic children's vocalizations and exploit the obtained knowledge to create advanced voice-based interfaces to enhance their therapeutic interventions. During the outgoing phase at MIT, the work will be about identifying and implementing machine learning algorithms for classifying children's vocalizations. The core strategy is to leverage the unique knowledge provided by caregivers who have long-term acquaintance with MV children with autism and can recognize the meaning of their vocalizations.

In the return phase at POLIMI, the project will be about designing, developing, and empirically validating a voice-based AAC prototype for children's speech therapy through a participatory-design process involving end users, their caregivers, and autism experts.

Who is involved

OrganisationCityCountryRole
Politecnico Di MilanoMilanoItalyCoordinator
Massachusetts Institute of TechnologyCambridgeUnited StatesAssociatedpartner

Source: CORDIS, European Commission. Data of 5 Nov 2026, reused under the Commission's reuse policy. View on CORDIS · Grant DOI