Researchers from Digital Media Solutions (DMS) Audio Lab within Samsung Research America received the Best Technical Paper Award during the conference at the 154th Audio Engineering Society (AES) Convention in Helsinki, Finland, in May 2023. AES Conventions are held annually in both the United States and Europe. They are the largest gatherings of researchers in signal processing for audio/acoustics, auditory perception, and machine learning worldwide. Several thousand attendees from academia and industry participate in a mutual exchange of new research in the area each year.

The paper resulted from the Visual Display (VD) TV and Audio products initiative to improve audio quality from consumer devices and involved stakeholders from the VD business unit and DMS Audio Lab researchers. DMS-Audio Lab is developing new technologies to push low-frequency performance, mainly bass, from thin TVs. "We are honored to receive this best paper award from AES. We have studied the state-of-the-art techniques and have shown that we outperform these in objective and subjective tests," said Allan Devantier, Vice President of R&D, DMS-Audio Lab.

This award paper, "Advances in Perceptual Bass Extension for Music and Cinematic Content," focuses on generating higher-frequency signals from the input audio stream that can be perceived as lower frequencies from devices that cannot reproduce these low-frequency signals. Traditional techniques introduce significant artifacts and include processing that cannot be well-controlled, leading to an unpredictable bass perception of music and cinematic content. To overcome these challenges, researchers from DMS-Audio proposed a new nonlinear synthesis function that had properties lacking in other approaches while maintaining good bass perception at arbitrary playback levels. Subjective and formal listening test results obtained from the listener pool confirmed the validity of the technique. New research and algorithms are in the works to further build on this work.

DMS-Audio Lab is located in Valencia, a suburb of Los Angeles, and the center of the world's music and entertainment (movie). The lab conducts research at the intersection of loudspeaker characterization and modeling, algorithms for signal processing towards audio/acoustics and machine learning, audio coding, and developing perceptual models of human hearing.