‌

CDT Cohort 1 student Dan Kumpik recently presented a poster at the UK Speech Conference 2022 which took place at the University of Edinburgh from 4-6th September 2022.

Here's what Dan had to say about the conference:

The conference was a research meeting for speech researchers, and I attended with two colleagues from a new research collaboration I am conducting with Southampton University on comparing human and machine perceptions of voice anonymity. Topics in talks and posters covered ranged from speech recognition systems to clinical speech interventions, including child speech therapy and detection of Parkinson’s disease from speech. I made LOADS of new connections- apart from meeting my collaborators in person for the first time, I also made lots of connections in the speech research community at Edinburgh University, and intend to leverage them if possible.

Dan adds that participating in the conference will help his PhD project in three ways:

Firstly, I have been exposed to valuable cutting-edge speech-specific research and new (to me) computational techniques that could be useful for my project. One good example was new approaches to speaker diarisation in noisy environments, which is the biggest challenge faced by my project, because our data consist of dementia patients talking while they watch TV. The second benefit is that I have made new connections which I hope to leverage for future collaborations. Thirdly, The strengthening of my existing collaboration on voice anonymisation suggests a possibility for using those techniques on my NHS dementia patients’ voices, which could be used for a chapter in my thesis.

This work is supported by the UK Engineering and Physical Sciences Research Council (EPSRC) through the Trustworthy Autonomous Systems Hub (EP/V00784X/1), a Turing AI Acceleration Fellowship on Citizen-Centric AI Systems  (EP/V022067/1) and the EPSRC Digital Health and Care Centre for Doctoral Training (CDT) at the University of Bristol (UKRI Grant No. EP/S023704/1).