[FFE] Merge 1.2.1-1 from Debian

Please merge 1.2.1-1 from Debian. I cannot work on this merge myself. This is only a reference.

Opus 1.2.1 has been in Debian since October 2017 and no new bugs were filed.

The current package version in Bionic 1.1.2 is the same as in Xenial. It's a pity that the improvements in the new version will not be present in the new LTS.

Looking at the upstream git log of opus [1], you'll see that there aren't (relatively) many commits after 1.2.1 up to this day, which means that 1.2.1 is a pretty stable release, and upstream goes directly to a major 1.3 release.
[1] https://git.xiph.org/?p=opus.git;a=shortlog

Version 1.2 brings improvements described here:
https://people.xiph.org/~jm/opus/opus-1.2/

From upstream News:
https://opus-codec.org/news/

libopus 1.2.1

Jun 26, 2017

This Opus 1.2.1 minor release fixes a relatively rare issue where the 1.2 encoder would wrongly assume a signal to be bandlimited to 12 kHz and not encode frequencies between 12 and 20 kHz. This only happens on a few clips, but it is good to update to avoid a potential loss of quality.

There are no other changes compared to 1.2. Please report any problems.

libopus 1.2

Jun 20, 2017

This Opus 1.2 major release brings many quality improvements, new features, and bug fixes. You can read all the details in this release demo page. Changes since 1.1.x include:

    Speech quality improvements especially in the 12-20 kbit/s range
    Improved VBR encoding for hybrid mode
    More aggressive use of wider speech bandwidth, including fullband speech starting at 14 kbit/s
    Music quality improvements in the 32-48 kb/s range
    Generic and SSE CELT optimizations
    Support for directly encoding packets up to 120 ms
    DTX support for CELT mode
    SILK CBR improvements
    Support for all of the fixes in draft-ietf-codec-opus-update-06 (the mono downmix and the folding fixes need --enable-update-draft)
    Many bug fixes, including integer wrap-arounds discovered through fuzzing (no security implications)

There are no known regressions compared to the latest stable release (1.1.5). Please report any problems.

libopus 1.2-rc1

Jun 8, 2017

Opus 1.2-rc1 is the first release candidate of the upcoming Opus 1.2 release. If no issues are found with it, it will soon become 1.2-final, so we encourage everyone to give it a try. Changes compared to 1.2-beta include:

    Improves quality on files with powerful tones that cause MDCT leakage
    Improves bit allocation on mode transitions (CELT to/from SILK/hybrid)
    More ARM Neon optimizations
    Fixes to the speech/music detection at the very beginning of files
    Fixes to the unit tests (fixes illegal instructions and –disable-static)

libopus 1.2-beta

May 24, 2017

This Opus 1.2-beta beta release of the upcoming Opus 1.2 fixes a bug in surround encoding causing very bad quality on signals beyond a certain amplitude. Thanks to Franziska Trojahn and others from HfT Leipzig for finding this issue through their listening test (the quality of the two bad files in the paper should now be similar to the others). Also included in this release are some more ARM Neon optimizations and some low-bitrate quality tuning.
----------------------------------------------------------------

(In case of not approving on updating to 1.2.1 from the Release Team, I suggest updating to the latest 1.1.x release, which is 1.1.5 [2], as it's a bug-fix release)
[2] https://git.xiph.org/?p=opus.git;a=shortlog;h=refs/heads/1.1.x
------------------------------------------------------------------

Changelog from Debian:

opus (1.2.1-1) unstable; urgency=medium

  * Fixes a bug in surround encoding causing very bad quality on loud signals
    beyond a certain amplitude.
  * Speech quality improvements especially in the 12-24 kbit/s range.
  * Music quality improvements in the 32-48 kb/s range.
  * More aggressive use of wider speech bandwidth, including fullband speech
    starting at 14 kbit/s.
  * Improved VBR encoding for hybrid mode.
  * SILK CBR improvements.
  * DTX support for CELT mode.
  * Support for directly encoding packets up to 120 ms.
  * Improves quality on files with powerful tones that cause MDCT leakage.
  * Fixes for speech/music detection at the very beginning of files.
  * Fixes an issue where the encoder can misdetect that the signal is SWB
    instead of FB, lowpassing the signal.

 -- Ron Lee <email address hidden> Thu, 28 Sep 2017 01:48:45 +0930

opus (1.2~alpha2-1) unstable; urgency=medium

  * Run the tonality analysis at 24 kHz, which reduces complexity while giving
    better frequency resolution for the tonality estimate.
  * Speech quality improvements especially in the 12-20 kbit/s range.
  * Improved VBR encoding for hybrid mode.
  * More aggressive use of wider speech bandwidth, including fullband speech
    starting at 14 kbit/s.
  * Music quality improvements in the 32-48 kb/s range.
  * Generic and SSE CELT optimizations.
  * Support for directly encoding packets up to 120 ms.
  * DTX support for CELT mode.
  * SILK CBR improvements.
  * Ensure that NLSF cannot be negative when computing a min distance between
    them. This was reported and fixed in July, and assessed as having only a
    relatively minor impact (garbage output, from the garbage input needed to
    trigger it), or at very worst, an assertion failure or simple crash from
    a slightly out of bounds read. In December it was assigned CVE-2017-0381
    by someone other than the upstream developers, with claims of it being a
    'Critical' issue on Android, but we're yet to see any analysis to back
    that up. Closes: #851612

 -- Ron Lee <email address hidden> Fri, 20 Jan 2017 02:48:31 +1030

opus (1.1.3-1) unstable; urgency=medium

  * Remove the generation date from the docs to improve reproducible builds.
    Closes: #825345
  * Fixes some corner cases with the soft clipper.
  * Improve CBR and some other issues seen with using crazy low rates.
  * More robustness fixes for reporting bad input.
  * More optimisations, mostly for ARM.
  * More precision fixes for edge cases.
  * Fixes to comfort noise generation.

 -- Ron Lee <email address hidden> Thu, 27 Oct 2016 23:58:06 +1030