SP.72 - Learning about perception of temporal fine structure by building audio codecs
3:20 PM - 5:20 PM
Thu Aug 22, 2019


Lars Villemoes* – Dolby Sweden AB, Stockholm, Sweden
Arijit Biswas Dolby Germany GmbH, Nürnberg, Germany
Heiko Purnhagen, Heidi-Maria Lehtonen Dolby Sweden AB, Stockholm, Sweden

The goal of audio coding is to efficiently describe an auditory experience while enabling a faithful reconstruction to the listener. The subjective quality compared to the original is measured by established psychoacoustic tests (BS.1116, BS.1534) and the description cost is measured in number of bits. As it is much cheaper to describe coarse scale signal properties than Temporal Fine Structure (TFS), tools like noise fill, spectral extension, binaural cue coding, and machine learning have increased performance of audio codecs far beyond the first generation based on masking principles (e.g. mp3). In this evolution, implicit knowledge on hearing has been acquired by codec developers, but it has become increasingly difficult to construct tools to predict subjective quality. For example, we still don't know which aspects of the TFS that are essential for the listening impression to be preserved. To explore this issue, we perform controlled experiments with models of auditory representations with the mindset from audio coding. As in the research on summary statistics of audio textures, the goal of the synthesis is to solve the inverse problem of creating a signal with a specified representation. However, our focus is on TFS aspects of nonstationary signals, and we show how evaluating a representation by listening immediately reveals strengths and weaknesses of candidate models.

Corresponding author: Lars Villemoes (

Lars Villemoes Dolby Sweden AB

Add to my calendar


Create your personal schedule through the official app, Whova!

Get Started