Add model card and links to paper/code

README.md

@@ -0,0 +1,35 @@
+---
+pipeline_tag: audio-classification
+---
+
+# MultiAPI-Spoof: Nes2Net-LA
+
+This repository contains the weights for **Nes2Net-LA**, a local-attention enhanced network for speech anti-spoofing detection and API tracing, as presented in the paper [MultiAPI Spoof: A Multi-API Dataset and Local-Attention Network for Speech Anti-spoofing Detection](https://arxiv.org/abs/2512.07352).
+
+- **Paper:** [https://arxiv.org/abs/2512.07352](https://arxiv.org/abs/2512.07352)
+- **Code:** [https://github.com/XuepingZhang/MultiAPI-Spoof](https://github.com/XuepingZhang/MultiAPI-Spoof)
+- **Project Page:** [https://xuepingzhang.github.io/MultiAPI-Spoof-Dataset/](https://xuepingzhang.github.io/MultiAPI-Spoof-Dataset/)
+
+## Overview
+
+Existing speech anti-spoofing benchmarks often rely on a narrow set of public models. **MultiAPI Spoof** addresses this gap by introducing a multi-API audio anti-spoofing dataset comprising about 230 hours of synthetic speech generated by 30 distinct APIs, including commercial services, open-source models, and online platforms.
+
+**Nes2Net-LA** is a local-attention enhanced variant of Nes2Net that improves local context modeling and fine-grained spoofing feature extraction. The model is designed for two primary tasks:
+1. **Anti-Spoofing Detection:** Classifying audio as bona fide or spoofed.
+2. **API Tracing:** Enabling fine-grained attribution of spoofed audio to its generation source.
+
+## Citation
+
+If you use this model or the MultiAPI-Spoof dataset, please cite the following work:
+
+```bibtex
+@misc{zhang2025multiapispoofmultiapidataset,
+      title={MultiAPI Spoof: A Multi-API Dataset and Local-Attention Network for Speech Anti-spoofing Detection}, 
+      author={Xueping Zhang and Zhenshan Zhang and Yechen Wang and Linxi Li and Liwei Jin and Ming Li},
+      year={2025},
+      eprint={2512.07352},
+      archivePrefix={arXiv},
+      primaryClass={cs.SD},
+      url={https://arxiv.org/abs/2512.07352}, 
+}
+```