docs: rewrite README to plan v3 format

README.md

@@ -1,160 +1,107 @@
 ---
 license: other
 language:
-- en
+  - en
+pretty_name: ASVspoof 2019 LA
 task_categories:
-- audio-classification
-pretty_name: ASVspoof2019LA
+  - audio-classification
 size_categories:
-- 10K<n<100K
+  - 10K<n<100K
 configs:
-- config_name: default
-  data_files:
-  - split: test
-    path: data/test-*.parquet
+  - config_name: default
+    data_files:
+      - split: test
+        path: "data/test-*.parquet"
 tags:
-- benchmark
-- audio
-- speech
-- spoofing-detection
-- deepfake-detection
-- anti-spoofing
-- logical-access
-- ASVspoof
-- arxiv:1911.01601
+  - anti-spoofing
+  - audio-deepfake-detection
+  - speech
+  - benchmark
+  - arena-ready
 paperswithcode_id: asvspoof-2019
+arxiv:
+  - "1911.01601"
 ---
 
-# ASVspoof2019LA
+# ASVspoof 2019 LA
 
-ASVspoof2019LA is a benchmark-ready packaging of the **Logical Access (LA) evaluation partition** from ASVspoof 2019 for speech spoofing and synthetic voice detection research.
+Benchmark-ready packaging of the **Logical Access (LA) evaluation partition** from ASVspoof 2019 for speech anti-spoofing and synthetic voice detection.
 
-This repository is intended for Hugging Face hosting and Papers with Code style benchmark tracking. It contains only the LA evaluation audio and the official LA countermeasure evaluation protocol. The original dataset at https://www.asvspoof.org/index2019.html was not modified.
+## Overview
 
-Contact: k.n.borodin@mtuci.ru
+This dataset contains the LA evaluation subset of the ASVspoof 2019 challenge. The task is binary classification: **bonafide** (genuine human speech) vs. **spoof** (synthetic or converted speech). The original dataset is available at https://www.asvspoof.org/index2019.html.
 
-## Contents
+## License & redistribution
 
-```text
-ASVspoof2019LA/
-  data/
-    test-00000-of-00009.parquet
-    test-00001-of-00009.parquet
-    ...
-    test-00008-of-00009.parquet
-  protocols/
-    ASVspoof2019.LA.cm.eval.trl.txt
-  submissions/
-    README.md
-    results_template.csv
-  build_parquet.py
-  eval.yaml
-  LICENSE.txt
-  README.md
-```
-
-## Benchmark Task
-
-Given an evaluation utterance from the ASVspoof 2019 Logical Access condition, predict whether it is:
-
-- `bonafide`: genuine human speech
-- `spoof`: synthetic or converted speech
-
-The benchmark uses the official ASVspoof 2019 LA countermeasure evaluation protocol:
-
-```text
-protocols/ASVspoof2019.LA.cm.eval.trl.txt
-```
-
-Protocol columns:
-
-```text
-speaker_id utterance_id unused attack_id label
-```
-
-Example:
+This dataset is redistributed under the **Open Data Commons Attribution License (ODC-By)**. See `LICENSE.txt` for the full text. The original dataset was not modified.
 
-```text
-LA_0039 LA_E_2834763 - A11 spoof
-LA_0004 LA_E_1665632 - - bonafide
-```
-
-Number of evaluation trials: **71,237**.
-
-## Metrics
-
-The leaderboard tracks **Equal Error Rate (EER %)** — lower is better.
-
-This is the primary metric on the Hugging Face benchmark leaderboard (task `eer`). EER is computed over all 71,237 evaluation trials from the official ASVspoof 2019 LA countermeasure protocol.
-
-When possible, also report **minimum tandem Detection Cost Function (min t-DCF)** in your paper alongside EER, using the official ASVspoof 2019 scoring implementation.
-
-## Dataset Schema
-
-Each row in the parquet files has the following columns:
+## Schema
 
 | Column | Type | Description |
 |--------|------|-------------|
-| `trial_id` | `string` | Utterance ID (unique key) |
-| `audio` | `Audio(16000)` | Audio waveform, auto-decoded at 16kHz |
-| `utterance_id` | `string` | Utterance identifier |
-| `speaker_id` | `string` | Speaker identifier |
-| `attack_id` | `string` | Attack type (`"-"` for bonafide) |
-| `label` | `ClassLabel` | `"bonafide"` or `"spoof"` |
-
-Label distribution: 7,355 bonafide, 63,882 spoof (71,237 total).
+| `path` | `string` | Stable archive-relative path (e.g. `LA_E_2834763.flac`), unique within dataset |
+| `audio` | `Audio(16000)` | Audio waveform, 16 kHz mono |
+| `label` | `ClassLabel` | `"bonafide"` (index 0) or `"spoof"` (index 1) |
+| `notes` | `string` | JSON with `utterance_id`, `speaker_id`, `subset` |
+
+`notes` example:
+```json
+{"utterance_id": "LA_E_2834763", "speaker_id": "LA_0039", "subset": "eval"}
+```
 
-## How to Use
+## Quick Start
 
 ```python
 from datasets import load_dataset
 
 ds = load_dataset("SpeechAntiSpoofingBenchmarks/ASVspoof2019_LA", split="test")
 print(ds[0])
-# {'trial_id': 'LA_E_...', 'audio': {'array': ..., 'sampling_rate': 16000}, ...}
+# {'path': 'LA_E_2834763.flac', 'audio': {'array': ..., 'sampling_rate': 16000},
+#  'label': 1, 'notes': '{"utterance_id": "LA_E_2834763", ...}'}
 ```
 
-## Submitting Results
+## Stats
 
-### Via Hugging Face (recommended)
+| Stat | Value |
+|------|-------|
+| Total trials | 71,237 |
+| Bonafide | 7,355 |
+| Spoof | 63,882 |
 
-To submit results to the benchmark leaderboard, add a YAML file to your model repo under `.eval_results/`:
+## Source provenance
 
-```yaml
-- dataset:
-    id: SpeechAntiSpoofingBenchmarks/ASVspoof2019_LA
-    task_id: eer
-  value: 0.83
-  date: "2026-05-19"
-  source:
-    url: https://github.com/clovaai/aasist
-    name: AASIST reference implementation
-  notes: "EER (%) on official LA eval protocol"
-```
+- Original dataset: https://www.asvspoof.org/index2019.html
+- Evaluation protocol: `protocols/ASVspoof2019.LA.cm.eval.trl.txt`
 
-Fields:
-- `dataset.id` — must be `SpeechAntiSpoofingBenchmarks/ASVspoof2019_LA`
-- `task_id` — must be `eer`
-- `value` — EER in percent (e.g. `0.83` means 0.83% EER)
-- `revision` — optional, pins the dataset snapshot for reproducibility
+## Evaluation
 
-### Via pull request
+For evaluation instructions and submission format, see [`submissions/README.md`](submissions/README.md).
 
-Alternatively, evaluate your system on the protocol in `protocols/ASVspoof2019.LA.cm.eval.trl.txt`, compute EER, and open a pull request adding results to `submissions/`.
+## Citation
 
-## Hugging Face Upload Notes
+**Original paper**: https://arxiv.org/abs/1911.01601
 
-Large files (parquet shards) are configured for Git LFS via `.gitattributes`.
+arXiv version:
 
-## Citation
+```bibtex
+@misc{wang2020asvspoof2019largescalepublic,
+      title={ASVspoof 2019: A large-scale public database of synthesized, converted and replayed speech},
+      author={Xin Wang and Junichi Yamagishi and Massimiliano Todisco and Hector Delgado and Andreas Nautsch and Nicholas Evans and Md Sahidullah and Ville Vestman and Tomi Kinnunen and Kong Aik Lee and Lauri Juvela and Paavo Alku and Yu-Huai Peng and Hsin-Te Hwang and Yu Tsao and Hsin-Min Wang and Sebastien Le Maguer and Markus Becker and Fergus Henderson and Rob Clark and Yu Zhang and Quan Wang and Ye Jia and Kai Onuma and Koji Mushika and Takashi Kaneda and Yuan Jiang and Li-Juan Liu and Yi-Chiao Wu and Wen-Chin Huang and Tomoki Toda and Kou Tanaka and Hirokazu Kameoka and Ingmar Steiner and Driss Matrouf and Jean-Francois Bonastre and Avashna Govender and Srikanth Ronanki and Jing-Xuan Zhang and Zhen-Hua Ling},
+      year={2020},
+      eprint={1911.01601},
+      archivePrefix={arXiv},
+      primaryClass={eess.AS},
+      url={https://arxiv.org/abs/1911.01601},
+}
+```
 
-If you use this benchmark, cite the original ASVspoof 2019 database paper:
+Peer-reviewed publication (Computer Speech & Language, 2020):
 
 ```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, Hector 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},
+  journal={Computer Speech \& Language},
   volume={64},
   pages={101114},
   year={2020},
@@ -162,18 +109,6 @@ If you use this benchmark, cite the original ASVspoof 2019 database paper:
 }
 ```
 
-```bibtex
-@misc{wang2020asvspoof2019largescalepublic,
-      title={ASVspoof 2019: A large-scale public database of synthesized, converted and replayed speech}, 
-      author={Xin Wang and Junichi Yamagishi and Massimiliano Todisco and Hector Delgado and Andreas Nautsch and Nicholas Evans and Md Sahidullah and Ville Vestman and Tomi Kinnunen and Kong Aik Lee and Lauri Juvela and Paavo Alku and Yu-Huai Peng and Hsin-Te Hwang and Yu Tsao and Hsin-Min Wang and Sebastien Le Maguer and Markus Becker and Fergus Henderson and Rob Clark and Yu Zhang and Quan Wang and Ye Jia and Kai Onuma and Koji Mushika and Takashi Kaneda and Yuan Jiang and Li-Juan Liu and Yi-Chiao Wu and Wen-Chin Huang and Tomoki Toda and Kou Tanaka and Hirokazu Kameoka and Ingmar Steiner and Driss Matrouf and Jean-Francois Bonastre and Avashna Govender and Srikanth Ronanki and Jing-Xuan Zhang and Zhen-Hua Ling},
-      year={2020},
-      eprint={1911.01601},
-      archivePrefix={arXiv},
-      primaryClass={eess.AS},
-      url={https://arxiv.org/abs/1911.01601}, 
-}
-```
-
-## License
+## Maintainer
 
-This benchmark package includes `LICENSE.txt` copied from the original ASVspoof 2019 LA distribution. Users are responsible for following the original dataset license and citation requirements.
+Contact: k.n.borodin@mtici.ru