Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Alex B]

I just stumbled on this:

From: http://www.macsimumnews.com/index.php/archive/apple_granted_new_patents_and_files_for_macexpo_trademark_in_china

Patent 7,072,477: Method and apparatus for automatically normalizing a perceived volume level in a digitally encoded file

This invention relates generally to multimedia devices. More specifically, the invention describes a method and apparatus that automatically normalizes the volume of a recording based upon both the selectivity and the sensitivity of the human auditory system.

Abstract: Automatically normalizing a perceived loudness for a digitally encoded audio track formed of a number of channels during playback on a multimedia asset player is described. A number of auditory selectivity frequency bands are selected and for each channel in the track, a power value for each of the number of selectivity frequency bands is computed. Each of the power values is weighted by a sensitivity weighting factor and a sum value of all the weighted power values is then calculated. For the track, a perceived acoustic power value is calculated based upon the sum value for each of the channels and a normalization gain factor based upon the perceived acoustic power is calculated and associated with the track. During playback, the normalization gain factor is applied to the track.

Patent point: Any track that is going to have its playback volume adjusted (i.e., normalized) has associated with it a perceived acoustic power value that characterizes the track as a whole. By automatically adjusting the player’s preamp gain to correct each track’s characteristic power, perceptually uniform average power (and therefore, loudness) across multiple tracks is achieved. It should be noted that characteristic power is an inherent property of a track and is essentially independent of the track’s particular encoding parameters (sampling frequency, bit rate, stereo mode, etc.). In a contemplated implementation, the perceived acoustic power value is pre-computed by a computing system separate and distinct from the multimedia asset player in those situations where the player has limited computational resources. Once computed, the perceived acoustic power value is stored in a database into which the track is stored in the form of, for example, an id tag associated with the track itself. Using the example, of the IPOD player, the perceived acoustic power value for a particular track is calculated as part of the ITUNES [TM.] software and stored with the digitally encoded track as part of a music library. In this arrangement, the perceived acoustic power value can be calculated as part of the background process that runs when the tracks are added to the music library. In this way, the computational resources required for normalization at playback is minimal requiring only a gain adjustment (i.e., a multiplication operation) and, if desired, a simple peak limiting to guard against over amplification. In addition, since the single gain adjustment applies for the duration of the entire track, there are no associated artifacts such as “pumping” caused by dynamically altering gain during playback of a track.

Inventor: William S. Kincaid

It seems that Apple has got a patent for a similar system that has been used e.g. in JRiver's programs since late 2001 (wasn't Replay Gain introduced in MJ8 beta?).

David Robinson's original proposal was published July 10th 2001 (http://www.replaygain.org/). Apple released its "Sound Check" feature in iTunes 3.0 about a year later (http://en.wikipedia.org/wiki/ITunes_version_history).

Perhaps it would be better to not change the name to "Volume Leveling" in MC12 since "Replay Gain" has provably been around longer than "Sound Check".