13 Commits
Author SHA1 Message Date
Juan Coria 9178d88b87 feat(core): allow loading Calibration object from an in-memory dict directly (#1883) 2025-06-13 13:40:05 +02:00
Juan Coria 799fffd768 improve(utils): add option to load model checkpoint from BytesIO (#1873) 2025-05-21 14:23:46 +02:00
Juan Coria 2ea4ea4870 fix: fix device mismatch in speechbrain embeddings 2022-07-02 11:36:45 +02:00
Juan Coria e4298a43a1 feat: add support for UEM 2022-01-28 13:56:36 +01:00
Juanma CoriaandHervé BREDIN eb39c87265 fix: do not reset task-dependent layers weights when changing task
Co-authored-by: Hervé BREDIN <hbredin@users.noreply.github.com>
2021-12-03 14:33:26 +01:00
Juanma Coria ab1205f43f feat: use AdaCos trick to compute scaling default value 2020-04-30 17:57:07 +02:00
Juanma Coria edc8830b57 feat: add support for center loss fine-tuning
fix: fix classification layer crash
2020-03-02 13:54:40 +01:00
Juanma Coria 8a06c02a3a feat: add --debug option to use PyTorch's anomaly detection 2020-02-17 12:02:11 +01:00
Juanma Coria a94b743df1 fix: fix naming compatibility issue 2020-01-30 15:27:05 +01:00
Juanma Coria 20e7f35cf4 feat: add x-vector architecture 2020-01-30 15:08:35 +01:00
Juanma Coria f43c34f918 fix: fix corner case where a file has no speech at all 2020-01-28 17:55:41 +01:00
Juanma Coria b64abc2804 feat: add temporal pooling layers in new pyannote.audio.models.pooling module 2020-01-28 17:00:42 +01:00
Juanma Coria d76e149b42 feat: add "bias" option to Classification approach 2020-01-24 14:01:25 +01:00