Difference between revisions of "Opus-1.1-alpha"

From XiphWiki
Jump to navigation Jump to search
m
 
(One intermediate revision by one other user not shown)
Line 11: Line 11:
 
* More aggressive use of dynamic allocation
 
* More aggressive use of dynamic allocation
 
* Automatic audio bandwidth detection
 
* Automatic audio bandwidth detection
 +
* DC rejection filter
  
 
== Optimizations ==
 
== Optimizations ==
Line 16: Line 17:
 
* Significant reduction of the decoder stack size requirement
 
* Significant reduction of the decoder stack size requirement
 
* Optimization of several functions for the common case of zero
 
* Optimization of several functions for the common case of zero
 +
 +
[[Category:Opus]]

Latest revision as of 07:31, 18 August 2015

This is an alpha release for the upcoming 1.1 release. Compared to 1.0.2, it includes quality improvements, optimizations, bug fixes, as well as an experimental speech/music detector for mode decisions. All the fixes and improvements from 1.0.2 are also in this release.

Quality

  • Unconstrained VBR algorithm
  • New transient detector
  • Bitrate boost for tonal samples, which are harder to code
  • Fixes to the tf analysis code
  • Improvements to the hybrid mode code
  • Fixed some mode switching issues
  • More aggressive use of dynamic allocation
  • Automatic audio bandwidth detection
  • DC rejection filter

Optimizations

  • Significant reduction of the decoder stack size requirement
  • Optimization of several functions for the common case of zero