How AURES works

Thanks to audio fingerprinting, a technology that uses fingerprints generated from an audio signal to identify and locate an audio sample in a library integrated in the player.  Unlike voice recognition, AURES is able to pick up not only the human voice but any sound – including music, ambient noises, background sound – and is therefore able to “listen” to the film as if it were a human ear. Nomen omen - from the Latin auris, is “ear”, but as numerous field tests have shown, unlike the human organ it is more accurate and reliable.

  • Where do fingerprints come from?

AURES generates them, via an integrated standalone tool, starting with the video on which the subtitles have been instantaneously synchronized. This guarantees system stability and avoids any compatibility problems at their root cause, while at the same time offering many other advantages such as:

  • “Always listening”

With a refresh rate of 2-3 seconds, AURES is in a “continuous listening” position, allowing it to match the film audio instantly and (re)positioning itself correctly in any scenario.

  • “Always at the right speed”

If the instantaneous subtitling work was carried out on a video file with a different frame rate to that of the film being shown in the theater, the subtitles will be out of sync.  To avoid this, a mechanism was introduced into the AURES audio fingerprinting library for detecting and adjusting any frame rate discrepancies.  No human intervention is therefore required, and this applies to any film and in any language.