Seed random-baseline submission for DFADD (#1)

- Seed random-baseline submission for DFADD (defda63488f4b4ef088210a6606ecf3613e0fe20)
- Fill reproduction block (match: scoring) (0f493bcfe72db1a8a5bc1db27d0d9d6bce7c5fdf)

submissions/random-baseline.yaml

@@ -0,0 +1,65 @@
+schema_version: 4
+
+system:
+  name: random-baseline
+  slug: random-baseline
+  description: >
+    Reference random baseline. Returns N(0, 1) for every utterance using a
+    fixed seed (seed=0). EER ≈ 50% by construction. Used as the seeded
+    smoke-test baseline for the arena (roadmap Phase 3).
+  code: https://github.com/SpeechAntiSpoofingBenchmarks/speech-spoof-bench
+  checkpoint: https://huggingface.co/SpeechAntiSpoofingBenchmarks/random-baseline-asas
+  params_millions: 1
+  paper:
+    arxiv_id: "1911.01601"
+    url: https://arxiv.org/abs/1911.01601
+    bibtex: |
+      @article{wang2020asvspoof,
+        title={ASVspoof 2019: A large-scale public database of synthesized,
+               converted and replayed speech},
+        author={Wang, Xin and Yamagishi, Junichi and Todisco, Massimiliano and
+                Delgado, H{\'e}ctor and Nautsch, Andreas and Evans, Nicholas
+                and Sahidullah, Md and Vestman, Ville and Kinnunen, Tomi and
+                Lee, Kong Aik and others},
+        journal={Computer Speech \& Language},
+        volume={64},
+        pages={101114},
+        year={2020},
+        publisher={Elsevier}
+      }
+
+dataset:
+  id: SpeechAntiSpoofingBenchmarks/DFADD
+  revision: c578c836da3b522b27d3dd85f89309f1737e5d31
+  split: test
+
+scores:
+  eer_percent: 48.21192052980132
+  n_trials: 3755
+  n_skipped: 0
+
+artifact:
+  scores_url: https://huggingface.co/SpeechAntiSpoofingBenchmarks/random-baseline-asas/resolve/78d28e1b7be5df39771b8f0f0151c68def153d58/.eval_results/SpeechAntiSpoofingBenchmarks/DFADD/scores.txt
+  scores_sha256: e217aa5e91d2410fdb9baeb5de7a291aeaa3d844efaa472fece67edc8a214bc8
+  bench_version: speech-spoof-bench==0.3.4
+
+reproduction:
+  reproduced_by: SpeechAntiSpoofingBenchmarks
+  reproduced_at: 2026-06-10
+  reproduced_bench_version: speech-spoof-bench==0.3.4
+  match: scoring
+
+submitter:
+  hf_username: SpeechAntiSpoofingBenchmarks
+  contact: k.n.borodin@mtuci.ru
+
+submitted_at: 2026-06-10
+
+notes: >
+  Seeded random baseline for the newly-added DFADD dataset (DFADD: The Diffusion
+  and Flow-Matching Based Audio Deepfake Dataset, arXiv 2409.08731; MIT;
+  VCTK-derived; eval/test split). 3755 trials: 755 bonafide (VCTK) / 3000 spoof
+  from 5 modern TTS (Grad-TTS, Matcha-TTS, NaturalSpeech 2, PFlow-TTS, StyleTTS 2,
+  600 each). Random baseline has no paper of its own; the paper field cites the
+  ASVspoof 2019 dataset paper as a placeholder, per the Phase 3 convention and
+  consistent with the other random-baseline submissions (same system slug).