Speech Processing in Realistic Environments - SPIRE
The SPIRE workshop took place on January 9th 2016 immediately following and in close connection to the SPIN 2016 workshop on 7-8 January.
In everyday life, people listen to speech under a wide range of conditions that are “sub-optimal” relative to the controlled conditions of laboratory experiments. Although listeners often experience more than one adverse condition simultaneously (e.g., noise and visual distraction), classical research methods have traditionally only addressed adverse conditions individually. This has contributed to the fragmentation of speech communication research into numerous sub-disciplines that rarely interact. While each type of adverse condition can have important consequences on its own, it is often the combination of conditions that conspires to create serious communication challenges especially for elderly and hearing-impaired individuals.
In 2012, a Marie Curie Initial Training Network called Investigating Speech Processing in Realistic Environments (INSPIRE) was initiated with the aim of creating a community of researchers capable of exploiting synergies between the sub-disciplines of speech communication. The purpose of this workshop is to bring together researchers with common interests in human and automatic speech recognition in challenging conditions (e.g., under increased cognitive load, divided attention, environmental noise, accented speech, non-native knowledge, hearing impairment & hearing loss).
Relevant topics include, but are not necessarily limited to:
- State of the art empirical research on speech perception in challenging, realistic listening environments
- Experimental and clinical methods for research in naturalistic speech perception
- Computational modeling of speech intelligibility for normal-hearing and hearing-impaired listeners under realistic conditions
- Tools and corpora for testing and comparing speech intelligibility
- Integration of human auditory processing and machine speech recognition