Tanel Alumäe, Head of the Language Technology Laboratory of the Institute of software Science at Tallinn University of Technology, will give a venia legendi lecture on "Weakly supervised training for speaker and language recognition", with which he will apply for the position of Professor of Speech Technology. The lecture can be viewed again here.
Training voice-based speaker and spoken language identification models traditionally requires a large amount of training data. However, manual annotation of training data is time consuming and may require a lot of expertise. This lecture introduces how to train speaker and language identification models based on freely available data on the Internet by using the metadata that comes along.
With this lecture Tanel Alumäe is applying for the position of Professor of Speech Processing. The purpose of the position is to ensure the sustainability of research and teaching in the field of speech technology by the Tallinn University of Technology.
Tanel Alumäe is the head of Laboratory of Language Technology at Tallinn University of Technology (TalTech). He received his PhD degree from the same university in 2006. After that, he has worked in several renowned research teams, including LIMSI/CNRS, Aalto University and Raytheon BBN Technologies. His recent research has focused on practical approaches to low-resource speech and language processing.
Watch the lecture again here. The lecture will be held in English.