AISSPER

Effective modeling of variabilities, exhibited by human’s speech at different levels (prosodic, acoustic, lexical,…), is still an open research problem for SLU systems. Even more, extracting relevant keywords, themes or concept mentions from either a sentence or an entire spoken document is a difficult task even for the most advanced end-to-end systems. AISSPER will propose new neural algorithms that will combine this multi-level information. For this purpose, AISSPER brings together renowned researchers in the field of artificial intelligence and automatic language processing. The teams immediately involved come from LIUM, UAPV and Orkis (Enterprise), and will benefit from the ongoing collaboration between LIUM and the Montreal Institute of Learning Algorithm within the European/Canadian M2CR project on neural architectures for multimodal and multilingual data processing.