Home


SAIL develops inclusive technologies and technologies for inclusion

 

 

SAIL focuses on human-centered signal & information processing that address key societal needs. Bridging science and engineering, SAILers pioneer behavioral signal processing and behavioral machine intelligence, affective computing, multimodal signal processing, computational media intelligence and computational speech science.

SAIL enables these through fundamental advances in audio, speech, language, image, video and bio signal processing, human and environment sensing and imaging, human-centered machine learning as well as applications developing speech, human language technologies, conversational and multimodal systems.

SAIL was established in 2000 by Professor Shrikanth (Shri) Narayanan and has made sustained contributions to numerous award winning papers, widely-used community resources, and broad impact through media mentions.

SAIL acknowledges sustained support from federal agencies (NSF, NIH, DARPA, IARPA, DoD, ONR, DoJ), foundations as well as industry grants and contracts.

.

Research Areas

 


Computational Speech Science: SPAN - Speech Production and Articulation Knowledge

Speech imaging, magnetic resonance imaging for dynamic vocal tract shaping and anatomy, understanding individual variability, speech development in children, prosody, articulatory-acoustic modeling, vocal song production and singing, clinical applications including neurological disorders and cancer

 

SLP: Speech and Language Processing Technologies

All aspects of the speech analysis/speech processing pipeline, speech and speaker recognition, diarization, spoken language understanding, speech emotion recognition, multilingual speech and language processing, applications in defense and intelligence, health and the arts

 

BioSP: Biosignal Sensing and Processing

Biosignal sensing and processing of physiological and physical activity, wearable sensing and mobile technologies, understanding mind-body-behavior connections, applications in health, work performance, learning

 

BSP: Behavioral Signal Processing

Computational methods for illuminating behavioral traits and mental states, tools for scientific discovery in behavioral phenotyping and behavior change modeling, applications in behavioral and mental health research and clinical translation (Addiction, Couple and Family relations, Autism Spectrum Disorders, Depression / Suicidality, Clinical encounters)

 

CMI: Computational Media Intelligence

Multimodal processing of audio (including music, speech, and environmental audio), image, video, and text, computational modeling of media content and its impact on individuals and society, applications in entertainment media (movies, TV , ads), news, social media, user-generated content

 

TMI: Trustworthy Machine Intelligence (Foundations of Signal Analysis and Interpretation)

Multimodal processing of audio (including music, speech, and environmental audio), image, video, and text, computational modeling of media content and its impact on individuals and society, applications in entertainment media (movies, TV , ads), news, social media, user-generated content

 

The picture above uses image resources from OpenMoji