Spatial audio coding by squeezing: analysis and application to compressing multiple soundfields

Spatially Squeezed Surround Audio Coding (S3AC) proposed by the authors provides efficient compression of multi-channel surround audio. Compression is achieved by exploiting human sound localisation blur to save the surround soundfield information in a squeezed stereo soundfield. In this paper, the localisation loss during the S3AC analysis/synthesis is evaluated and the minimum size of the S3AC squeezed soundfield is derived in a frequency dependent form. Results from perceptual listening tests show that, compared with standard squeezing from a 360° surround soundfield to a 60° stereo soundfield, a more intensive squeezing method, such as from 360°-to-5°, does not introduce audible localisation distortion. This leads to a further application of S3AC for compressing more than one surround soundfield into a single stereo downmix for applications such as spatialised teleconferencing. This application is also described and perceptually evaluate.