Archontis Politis, Leo McCormack, and Ville Pulkki

Enhancement of Ambisonic Binaural Reproduction using Directional Audio Coding with Optimal Adaptive Mixing

Companion page for a paper in the 2017 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, New York, USA, October 15–18, 2017


Headphone reproduction of recorded spatial sound scenes is of great interest in the field of immersive audiovisual tech- nologies. Directional Audio Coding (DirAC) is an established perceptually-motivated parametric spatial audio coding method that can achieve high-quality headphone reproduction, surpassing popu- lar non-parametric methods such as first-order Ambisonics (FOA). The earlier version of DirAC for headphones was limited to FOA input and achieved binaural rendering through a virtual loudspeaker approach, resulting in a high computational overhead. Therefore, this work proposes an improved DirAC method that directly syn- thesises the binaural cues based on the estimated spatial parameters. The method can accommodate higher-order Ambisonics (HOA) signals and has reduced computational requirements; thus, mak- ing it suitable for lightweight processing with fast update rates and head-tracking support. According to listening tests, using only FOA signals, the method results in equal or higher spatial accuracy to third-order Ambisonics and far surpasses FOA.


The paper can be found on the IEEE publications website.

Listening test sound scenes

The samples used for the test were first designed as 28-channel loudspeaker signals, by distributing anechoic signals to loudspeakers, while deliberately designing them to be critical of spatial reproduction methods. A mbisonic encoders were then applied to each of the 28 channel sound scenes, in order to obtain both FOA and third-order Ambisonic signals.

The reference test cases were obtained by convolv-ing the 28 loudspeaker signals with their respective HRTFs and summing the resulting binaural signals. The anchor was generated by extracting the omnidirectional signal from the FOA signals and panning it to the left and right channels equally. The test cases for both first and third order Ambisonics and OM-DirAC were obtained by passing the corresponding ambisonic signals through their respective off-line decoders.





Updated on Wednesday April 26, 2017
This page uses HTML5, CSS, and JavaScript