SPLANN

SOFTWIN announces the start of the project Automatic Infant Crying Recognition System (SPLANN), contract: 25/2014, Submission Code: PN-II-PT-PCCA-2013-4-1443.

SPLANN is an applied research project in the field of neonatology and pattern recognition, involving collaboration between a software development company, a university and a medical facility. The project falls into the "Information and communication technology" domain, as it involves processing the information inherent to the infant cry using electronic computing devices. The project falls within the research theme "Health and Environment Information Systems (eHealth)", the main objective of the project being the development of an automated (computerized) system for infant language recognition, applicable in the health domain, respectively neonatology and child care areas.
 
Project objectives

The infant cry represents an important instrument which can be used to diagnose different pathologies and to decipher his/her needs. Researches show that a consistent recognition of the infant’s state, in a short time, ensures an optimal physical and mental development for the infant, as well as significantly reduce the parents’ stress level. The SPLANN project seeks meeting the needs of families and aims the design and development of an intelligent system for automatic recognition of infant cry (language).

A main scientific objective of the project consists in establishing a correspondence between the sequences of cries made by an infant and his/hers state/need, by studying the physical model of producing crying and the significance of the newborn language. The project will use research performed in speech technology and medicine domains by the two partners: the Faculty of Electronics, Telecommunications and Information Technology and the Emergency Clinical Hospital "St. Pantelimon".

A second scientific objective refers to defining innovative algorithms for recognizing the infant language by adapting and optimizing a series of signal processing and pattern recognition algorithms, property of SOFTWIN. The algorithms are parts of several patents registered or in progress of registration, by the Coordinator of the project, at national and international level.

The achievement of the main objectives depends on a series of secondary objectives:

  • making a study of the correspondence between the sequences of sounds ( made by the infant and his/her needs;
  • creating a database of infant cries, database which will allow both the validation of the proposed methods, as well as further developments.

Together, the study and the database will establish the “ground truth” of the project.
The consortium will also make an Application Programming Interface (API) for the development of automatic infant language recognition solutions. The API will integrate a series of libraries specific to acquisition, processing and recognition modules. From the way the API will be developed, it will ease the integration on several platforms (PCs, mobile devices, dedicated devices), while maintaining the consistency of cry interpretation.

 

Acest site foloseşte cookie-uri. Folosim cookie-uri pentru analiza şi îmbunătăţirea site-ului, personalizarea vizitei, marketing şi reclamă. Prin navigarea pe acest site, vă exprimaţi acordul asupra folosirii cookie-urilor în aceste scopuri. Citiți mai mult.