Study of the Effects of Active Noise Cancellation on Music Playback

Published in Noise and Vibration Conference & Exhibition, SAE Technical Paper, 2021

Technology for automotive active noise cancellation (ANC) such as HARMAN’s Engine Order Cancellation (EOC) system help reduce in-cabin noise levels using a set of error microphones and the vehicle’s built-in audio system to generate anti-noise signals. A major benefit of these systems is that significant reduction in unwanted noise levels in the 20 - 400 Hz range can be achieved without the addition of extra noise control material. However, in the instances where the frequencies of anti-noise signals and music signals overlap, a degradation in the overall music reproduction quality is possible. In this paper, we study the effects of EOC on music playback by applying signal processing and statistical methods to objectively measure degradation in audio content when EOC is active. This study is carried out using production EOC software in a simulation environment. The simulation models the cabin acoustic response from measured vehicles and uses recordings of vehicle noise and music as inputs. For each test case, we simulate in-cabin music playback in the presence of engine noise with and without EOC active. This analysis is performed on a large set of music spanning various genres to maximize the likelihood of matching the listening preferences of our end-users. Finally, we also show that musical interference suppression (MIS) algorithms such as TrueAudio can reduce degradation in music reproduction quality when the frequencies of ANC anti-noise signals and music overlap.

Recommended citation: Basu, S.; Tackett, J., Trumpy, D.; Walt, A.; Adari, S. (2021). "Study of the Effects of Active Noise Cancellation on Music Playback." SAE Technical Paper.

Musical Polyphony Estimation

Published in Proceedings of the Audio Engineering Society, Convention 144, 2018

Knowing the number of sources present in a mixture is useful for many computer audition problems such as polyphonic music transcription, source separation, and speech enhancement. Most existing algorithms for these applications require the user to provide this number thereby limiting the possibility of complete automatization. In this paper we explore a few probabilistic and machine learning approaches for an autonomous source number estimation. We then propose an implementation of a multi-class classification method using convolutional neural networks for musical polyphony estimation. In addition, we use these results to improve the performance of an instrument classifier based on the same dataset. Our final classification results for both the networks, prove that this method is a promising starting point for further advancements in unsupervised source counting and separation algorithms for music and speech.

Recommended citation: S. Kareer, and S. Basu, "Musical Polyphony Estimation," Engineering Brief 434, (2018 May.)'

Bring a Concert Home

Published in Proceedings of the Audio Engineering Society, Convention 143, 2017

Inverse filtering of rooms to improve their frequency response or reverberation time is a well-researched topic in acoustical signal processing. With the aim of giving music lovers the experience of a concert hall in their own homes, we describe a system that employs signal processing techniques, including inverse filtering, to accurately reproduce concert hall acoustics in a home listening space. First, binaural impulse responses were measured at a few chosen seating positions in the concert hall. Next, the listening location along with its loudspeaker configuration is acoustically characterized and inverse filtered using MINT and Cross-talk Cancellation algorithms to produce a flat-frequency response. We observed that speech and music, after our inverse filtering method showed near-anechoic qualities which allowed us to subsequently impress the acoustical response of a wide range of concert halls upon the original audio. A demonstration will be provided using 4 loudspeakers for a quadraphonic sound reproduction at the listening area. In continuing work, to produce a sufficiently wide listening area, we are combining head tracking with adaptive inverse filtering to adjust to the listeners’ movements.

Recommended citation: S. Basu, and S. Kareer, "The BACH Experience: Bring a Concert Home," Engineering Brief 393, (2017 October.).