top of page

WIYE: building a corpus of children's audio and video recordings with a story-based app

WIYE: costruzione di un corpus di registrazioni audio e video dei bambini con un'app basata su storie

Spitale Micol, Catania Fabio, Cosentino Giulia, Gelsomini Mirko, Garzotto Franca

Venue:

Companion Proceedings of the 24th International Conference on Intelligent User Interfaces

Year:

2019

Type:

Conference Paper

URL:

Abstract (EN)

This paper describes the procedure to create an emotional data-set, named WIYE (What Is Your Emotion?), composed by semantic contents, audio and video recordings of children. Data have been collected using an interactive storytelling application, leading children aged 4 years to 12 years to discover about their emotional sphere and emotion expression skills. During the story, every episode is dedicated and focused on a specific emotion. The investigated emotional states are sadness, anger, fear, surprise and joy. This corpus can be exploited by Conversational Technologies: it can be used with Machine Learning classification algorithms to train models to recognize emotions expressed by children starting from the pitch of their voice, their facial expression and the contents of their conversations.

Abstract (IT)

Questo articolo descrive la procedura per creare un dataset emotivo, denominato WIYE (What Is Your Emotion?), composto da contenuti semantici, registrazioni audio e video di bambini. I dati sono stati raccolti utilizzando un'app interattiva di narrazione, che guida i bambini di età compresa tra 4 e 12 anni a esplorare la loro sfera emotiva e le capacità di espressione delle emozioni. Durante la storia, ogni episodio è dedicato a una specifica emozione, tra cui tristezza, rabbia, paura, sorpresa e gioia. Questo corpus può essere sfruttato dalle tecnologie conversazionali per addestrare modelli di riconoscimento delle emozioni basati sulla voce, l'espressione facciale e i contenuti delle conversazioni dei bambini.

Dreams are wishes... let's turn them into reality! Whether it's an innovative project, a collaboration, or simply a curiosity,
don't hesitate to write to us.

SENSEi - affirmation

Thanks for submitting!

The SENSEi platform is available as open-source for research purposes.
If you are an academic institution or researcher interested in integrating SENSEi into your studies, please contact us.

SENSEi - footer

® SENSEi | Registered Trademark EUIPO #018794538 | All rights reserved.
Intellectual Property owned by Mirko Gelsomini | VAT: IT10831300966

bottom of page