Views
No views yet
speech_enhancement DriftSE backend can pin a revision
and a content hash. Nothing here is original work of the Senseable Intelligence
Group.695a64db187500fa0d7bae23912680bd5d4df613.logs/distillhubert_three_layers_with_z/last.ckpt).config.json is upstream's config/with_z/v2_drift2_distillhubert_three_layers.json
at the same commit.LICENSE file and no licence statement, so the
default is all rights reserved. It is also built on
SGMSE+ (MIT) without carrying that statement
forward. A request for an explicit licence has been opened upstream; this mirror
stays private until it is answered. Do not make it public without that answer.| File | What it is |
|---|---|
last.ckpt | logs/distillhubert_three_layers_with_z/last.ckpt. Pairs with config.json. This is the variant senselab's backend pins. |
config.json | Upstream config/with_z/v2_drift2_distillhubert_three_layers.json. |
last_pesq_sisdr_ccmse.ckpt | logs/distillhubert_three_layers_pesq_sisdr_ccmse_with_z/last.ckpt — the DriftSE† row, jointly trained with auxiliary PESQ and SI-SDR losses, and the best published numbers on both benchmark tables (VB-DMD PESQ 3.45 / SI-SDR 20.60; DNS-2020 WV-MOS 2.65 / SCOREQ 2.97). Pairs with config_pesq_sisdr_ccmse.json. |
config_pesq_sisdr_ccmse.json | Upstream config/with_z/v2_drift2_distillhubert_three_layers_pesq_sisdr_ccmse.json. |
logs/ archive, which
holds seven variants in total; these two are the ones worth keeping. They are
mirrored verbatim, so each still contains its optimizer and scheduler
state — that is most of the ~1.1 GB. Stripping it would shrink the download but
would make these derived artifacts rather than a faithful mirror.latent_ckpt/
archive is not mirrored here.model vs ema. Each checkpoint contains both a model and an ema state
dict. Upstream's enhancement.py loads checkpoint["model"], so that is what
senselab loads and what reproduces the published numbers — even though EMA weights
usually score better, and even though the sibling unasdiff codebase does the
opposite and loads its ema copy. If you switch to ema, you are no longer
running what the paper reports; measure before believing it.1@inproceedings{xu2026driftse,
2 author = {Liang Xu and Diego Caviedes-Nozal and W. Bastiaan Kleijn and Longfei Felix Yan and Rasmus Kongsgaard Olsson},
3 title = {Speech Enhancement Based on Drifting Models},
4 booktitle = {Proc. Interspeech 2026},
5 year = {2026}
6}