Skip to content

Releases: hyperion-ml/hyperion

v0.3.2 Clotho Fix 2

25 Mar 16:38
c4c9eee
Compare
Choose a tag to compare
  • Fix bug in model loader for older x-vector models

v0.3.1 Clotho Fix 1

03 Mar 17:54
61cf1a4
Compare
Choose a tag to compare
  • Fix bug in model loader for older x-vector models
  • Fix bug in visual recipes
  • Adds some docs

v0.3.0 Clotho

03 Jan 15:11
4144d61
Compare
Choose a tag to compare
  • Breaking Changes:
    • s parameter in x-vector AAM/AM softmax loss is renamed as cos_scale to avoid conflict with other arguments that start with s. Models trained with older versions still should work.
    • Launcher for multigpu has been changed to the newer torchrun, it implies the multigpu won't work on PyTorch <1.9.
  • Hyperion updated to work with PyTorch >=1.9, PyTorch 1.10 recommended.
  • Documentation automatically uploaded to https://hyperion-ml.readthedocs.io

v0.2.6

24 Dec 18:46
bbebf45
Compare
Choose a tag to compare
  • This release, should upload automatically to pypi

v0.2.2

24 Dec 16:40
3f6463e
Compare
Choose a tag to compare
  • All Code formatted with Black
  • Added GitHub Action to check for black formatting

v0.2.1

22 Dec 13:13
7b0aeb0
Compare
Choose a tag to compare
  • Changes in requirements, current version only works with
    • Pytorch >=1.6 <= 1.8.0
    • Fairscale = 0.3.8

v0.2.0 Tyche

10 Dec 21:10
4bbc698
Compare
Choose a tag to compare
  • Adds ECAPA-TDNN
  • New Recipes:
    • SRE21-AV Audio: 16k and 8k versions
    • SRE21-AV Visual: face recognition recipes using Pre-trained MX-Net and Pytorch ArcFace embeddings
    • SRE21-AV: fusion of audio and visual modalities at score levels

v0.1.0 Lachesis

15 Aug 23:49
937c95c
Compare
Choose a tag to compare
  • Make Hyperion pip installable
  • Adds installation instructions
  • Configuration files and command-line arguments are handled using jsonarparse, not argparse anymore. This allows us to use yaml files and overrides the values in the yaml file from the command line.
  • First version using nn.DistributedDataParallel instead of nn.DataParallel
  • Supports FairScale Sharded DataParallel, we didn't observe significant memory gains in our models using this so far
  • Added SpineNet, Spine2Net and TSE-Spine2Net x-vector architectures from our IS21 paper
  • Added SpeAugment PyTorch Layer
  • Added numpy speed augment class
  • Fixed make_voxceleb2cat.pl, all speakers extracted from the same video were getting the same spkid, the script did not take into account that more than one speaker could be extracted from each video
  • New recipes:
    • Recipe for classifying adversarial attacks algorithms and threat models from our IS21 paper (voxceleb/adv.v2)
    • Recipe for adv attacks aginst spk verif renamed as voxceleb/adv.v1 and adv.v1.1, these recipes have been updated and cleaned up
    • Recipe for SRE19-AV Audio part with AHC diarization (sre19-av-a/v2.1)
    • Recipe for Chime5 speaker verification setup chime5/v1
    • Recipes for SRE19-AV Face Recognition using pretrained RetinaFace face detector and ArcFace embeddings from InsightFace MX-Net repository (sre19-av-v/v0.1) and Insightface-Pytorch (sre19-av-v/v0.2)
    • Added VOiCES challenge recipe
    • Adds SRE20-CTS recipe v1
    • Added Spine2Net results in voxceleb/v1.1 recipe

v0.1.0-rc.1

26 Jul 22:15
75eb740
Compare
Choose a tag to compare
v0.1.0-rc.1 Pre-release
Pre-release
  • Added Spec and speed augment

v0.1.0-beta

04 Jul 16:30
c74b412
Compare
Choose a tag to compare
v0.1.0-beta Pre-release
Pre-release
  • Make Hyperion pip installable
  • Adds installation instructions