Skip to main content

University of Salford home

Acoustics

  • Acoustics Research Centre
  • Commercial testing
  • Acoustic PhD opportunities
  • Research subject areas
  • Acoustics Research Staff
  • Acoustics Research Contacts
  • Seminars
  • Acoustics & Audio Taught Programme Employer and Alumni Surveys
  • Acoustics and Audio Courses
  1. Home
  2. Seminars
  3. Webinar 5/2/25: Leveraging deep learning to improve the video conferencing experience.

Webinar 5/2/25: Leveraging deep learning to improve the video conferencing experience.

Posted by: rebeccavos
Categories:
  • Seminars
Tags: acoustics, AI, audio, spatial audio, speech enhancement

Speaker: Dr. Paul Kendrick, Logitech

Wed 5th February 12:00 – 13:00 UK time. This seminar will be held in person in Gilbert 1 and you can join us online via Microsoft Teams.

image of Paul Kendrick

Abstract

Video conferencing has become an essential tool for communication in both professional and personal contexts, revolutionizing the way we connect and collaborate. However, delivering seamless and effective video conferencing experiences poses a range of technical challenges. At Logitech, deep learning has become a cornerstone of efforts to enhance these systems, driving innovations that improve audio and video quality in complex environments.

In this seminar, Dr. Paul Kendrick will provide an overview of the challenges associated with video conferencing, focusing on audio tasks such as speech enhancement and direction of arrival estimation, alongside the integration of audio and video modalities for tasks such as speaker localization and tracking. These tasks require solving unique problems in real-time, often in environments with noise, reverberation, and multiple simultaneous speakers.

Dr. Kendrick will also discuss the broader challenges of transitioning cutting-edge research into production-ready solutions. This includes the rigorous process of ensuring that new features are not only innovative but also robust, scalable, and serviceable in diverse real-world applications. The seminar will offer insights into the interdisciplinary approaches and practical considerations required to bridge the gap between research and product development, ultimately delivering technologies that enhance the user experience.

Biography

Dr. Paul Kendrick is a Senior Audio Machine Learning Engineer at Logitech, where he develops algorithms to improve teleconferencing experiences. His work focuses on enhancing the quality and effectiveness of virtual meetings, helping users connect and collaborate more seamlessly.

Dr. Kendrick earned a PhD in 2009 from the University of Salford, where his research centered on “Blind Estimation of Room Acoustic Parameters from Speech and Music Signals.” After completing his doctorate, he remained at Salford as a Lecturer and researcher, contributing to the fields of acoustics and signal processing.

In 2019, he moved into the audio technology industry, joining Music Tribe to work on intelligent mixing desks. This role allowed him to apply his academic expertise to practical audio solutions for musicians and producers. Since 2022, Dr. Kendrick has been with Logitech, where he continues to focus on improving audio experiences in teleconferencing.

Seminars

  • Webinar 7/5/25: Modal analysis of signal processing systems: applications for room acoustics
  • Webinar 30/04/25: Time domain vibro-acoustic reduced order models and their potential for digital twins
  • Webinar 23/04/25: Access as Artistic Resources

Contact Us

Head of Acoustics Research
Professor David Waddington
d.c.waddington@salford.ac.uk

© 2025 University of Salford

  • Acoustics on YouTube
  • Acoustics on Facebook
  • Acoustics on Instagram
  • Acoustics on LinkedIn