Add wav2vec2-large-antideepfake-nda submission for EmoFake_test (EER 0.71%, reproduced: scoring)

submissions/wav2vec2-large-antideepfake-nda.yaml

@@ -0,0 +1,53 @@
+schema_version: 4
+system:
+  name: Wav2Vec2-Large-AntiDeepfake-NDA
+  slug: wav2vec2-large-antideepfake-nda
+  description: 'AntiDeepfake detector (wav2vec 2.0 Large backbone): a self-supervised
+    speech foundation model post-trained for deepfake/spoof detection, with an AdaptiveAvgPool1d
+    + single fully-connected layer head over the SSL features. Released by nii-yamagishilab
+    (arXiv:2506.21090). Trained **without** data augmentation (NDA variant). Checkpoint
+    copied bit-exactly from nii-yamagishilab/wav2vec-large-anti-deepfake-nda. Evaluated
+    here with a deterministic first-4 s window and batching (NOT the paper''s whole-utterance
+    protocol).
+
+    '
+  code: https://github.com/nii-yamagishilab/AntiDeepfake
+  checkpoint: https://huggingface.co/SpeechAntiSpoofingBenchmarks/Wav2Vec2-Large-AntiDeepfake-NDA
+  paper:
+    arxiv_id: '2506.21090'
+    url: https://arxiv.org/abs/2506.21090
+    bibtex: "@article{ge2025posttraining,\n  title={Post-training for Deepfake Speech\
+      \ Detection},\n  author={Ge, Wanying and Wang, Xin and Kawahara, Tomoki and\
+      \ Yamamoto, Yuto and Delgado, Hector and Todisco, Massimiliano and Evans, Nicholas\
+      \ and Yamagishi, Junichi},\n  journal={arXiv preprint arXiv:2506.21090},\n \
+      \ year={2025}\n}\n"
+  params_millions: 315.0
+dataset:
+  id: SpeechAntiSpoofingBenchmarks/EmoFake_test
+  revision: 8e3fab1edce07654310631891d871824510c7945
+  split: test
+scores:
+  eer_percent: 0.7071428571428572
+  n_trials: 17500
+  n_skipped: 0
+artifact:
+  scores_url: https://huggingface.co/SpeechAntiSpoofingBenchmarks/Wav2Vec2-Large-AntiDeepfake-NDA/resolve/b923e129a57cd3547f1556156b074b850222873a/.eval_results/SpeechAntiSpoofingBenchmarks/EmoFake_test/scores.txt
+  scores_sha256: 12dae92c80b12374d685165e5c6e8a65db069b8bdba94c412a0245410d65607e
+  bench_version: speech-spoof-bench==0.4.1
+reproduction:
+  reproduced_by: SpeechAntiSpoofingBenchmarks
+  reproduced_at: '2026-06-24'
+  reproduced_bench_version: speech-spoof-bench==0.4.1
+  match: scoring
+submitter:
+  hf_username: korallll
+  contact: k.n.borodin@mtuci.ru
+submitted_at: '2026-06-24'
+notes: 'score = output logit for class 1 (real / bona fide); higher = more bona fide.
+  Backbone built from the source fairseq config (w2v_large); the released model.safetensors
+  overwrites every weight (220-key exact match family-wide). Input: fixed first 64000
+  samples (4.0 s @ 16 kHz), tile-repeated if shorter, per-utterance layer_norm; no
+  resampling. Checkpoint license: CC BY-NC-SA 4.0. STRICTLY PROHIBITED on the model''s
+  in-domain corpora (see README); Arena runs ArAD + In-the-Wild only.
+
+  '