What is Audile ?

Audile offers four main functionnalities :

• A coding/decoding and compact description for audio signals that allows for progressive processing and decoding. This provides a full level-of-detail like representation of audio signals.
• A dynamic auditory masking algorithm which uses the compact description of the audio signals to cull away inaudible sound sources at each time-frame of the simulation (typically every 20 ms.).
• A spatial coding algorithm that can represent the spatial information of all sound sources in the scene in a hierarchically. In particular groups of sound sources can be dynamically grouped and replaced by a unique “impostor” source with minimal degradation in spatial perception. This algorithm is very efficient so that the representation can be re-adapted at every frame of the simulation.
• An algorithm that decodes our progressively encoded signals taking into account their relative importance to the global auditory scene.  This is used to optimize processing and spatial audio algorithms applied to each sound source.

An example of the combined use of these four components is shown in the figure below.

However, it is of course possible to use them completely independently from each other.

AUDILE was developed by the REVES research group