Skip to content

Releases: gudgud96/frechet-audio-distance

v0.3.1

25 Mar 15:01
Compare
Choose a tag to compare

What's Changed

Several major features added in this release, including the EnCodec embedding support, MPS support, and CLAP text-audio score calculation.
Huge thanks to all contributors!

  • [feature] add EnCodec model embeddings to FAD calculation by @ivanlmh in #23
  • fix fad calculation for newer versions of scipy by @zhvng in #25
  • add automatic MPS device support by @balintlaczko in #27
  • [feature] add CLAP score metric to compute similarity between text and audio embeddings by @mcomunita in #28

New Contributors

Full Changelog: 0.2.0...0.3.0

v0.2.0

17 Oct 14:15
47dbd5c
Compare
Choose a tag to compare

What's Changed

Full Changelog: 0.1.2...0.2.0

v0.1.2

15 Aug 02:58
Compare
Choose a tag to compare

What's Changed

New Contributors

Full Changelog: 0.1.1...0.1.2

v0.1.1

17 Jul 03:57
a116459
Compare
Choose a tag to compare

What's Changed

Full Changelog: 0.1.0...0.1.1

v0.1.0

22 Apr 04:07
Compare
Choose a tag to compare

What's Changed

  • [fix] Fix stereo shape issue by converting to mono by @gudgud96 in #2
  • [chore] Add dependencies to pyproject.toml + add FAD comparison by @gudgud96 in #4
  • [chore] Add score results by @gudgud96 in #5
  • [refactor] Remove redundant fixed sample rate packing by @gudgud96 in #8
  • [feature] Add PANN model embedding to FAD calculation by @gudgud96 in #9
  • [refactor] Remove "list of audio paths" input type for get_embedding function by @gudgud96 in #10

New Contributors

Full Changelog: https://github.com/gudgud96/frechet-audio-distance/commits/0.1.0