Downloads · 30 days
0
ShiyiLIN0922/WildEar-HZ-strict-005
WildEar-HZ-strict-005 is a audio classification model from ShiyiLIN0922. Use it for the audio classification task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as apache-2.0.
Downloads · 30 days
0
Access
Public
Updated Aug 8, 2026
Repo size
—
Likes
0
Public
Click a slice to open those files.
.pdf2 MB · 91%
From the Hugging Face model README
中文在前,English follows.
WildEar-HZ strict-005 是一个轻量生物声学后训练包。它不包含 BioLingual 基础模型权重,而是包含:
运行时使用 davidrrobinson/BioLingual 冻结音频编码器生成 512 维单位嵌入。
候选内显示分数是温度归一化的相对分数,不是校准后的真实后验概率。
许可弱标签音频按上传者或来源组全局隔离。高文本 margin 窗口用于构建类别原型:
p_c = normalize(mean(z_i))
题目方明确标签声音构建参考原型:
p'_c = normalize((1 - alpha) p_c + alpha r_c)
alpha=0.15 仅由 validation 选择。21 类中有 13 类使用训练音频原型,8 类使用文本回退。240 条野外候选中的 16 条 AI 伪标签未通过固定 validation 门禁,生产采用数为 0。
范围:454 条录音、13 类、每类至少 5 个独立来源组。
| 方法 | Accuracy | Macro-F1 | 前三候选 |
|---|---|---|---|
| BioLingual 零样本 | 88.11% | 83.94% | 96.26% |
| 类别原型后训练 | 94.05% | 91.95% | 99.12% |
| Class-balanced 线性头 | 94.71% | 93.09% | 99.34% |
原型后训练相对零样本的上传者组 Bootstrap 提升:
线性头相对原型法的区间跨 0,因此当前证据不支持显著优势。生产版本保留类别原型。
非目标动物按上传者全局拆分;环境声按 ESC 原始 fold 拆分。非目标原型训练、阈值选择和 test 交集为 0。
| 分组 | AUROC | 线上 Known recall | 线上 FAR |
|---|---|---|---|
| 总体 | 0.9696 | 72.73% | 6.45% |
| 非目标动物 | 0.9500 | 72.73% | 20.00% |
| 环境声音 | 0.9789 | 72.73% | 0.00% |
Unknown 是保守风险提示,不替代专家物种鉴定。
wildear_biolingual_prototypes.npz:目标与文本原型;wildear_unknown_animal_prototypes.npz:非目标原型与线上阈值;wild-ear-hz-strict-005.json:release manifest 与 SHA-256;reports/:最终 test、嵌套分组 CV、独立 open-set test;TECHNICAL_REPORT_ZH_EN.md:中英双语技术报告;TECHNICAL_REPORT_ZH_EN.pdf:中英双语详细 PDF。WildEar-HZ strict-005 is a lightweight bioacoustic post-training package. It does not redistribute BioLingual base weights. It contains:
At runtime, the frozen davidrrobinson/BioLingual encoder produces normalized 512-dimensional audio embeddings.
Displayed candidate scores are temperature-normalized relative scores, not calibrated posterior probabilities.
Licensed weak-label audio is globally separated by uploader/source group. High text-margin windows form class prototypes:
p_c = normalize(mean(z_i))
Explicit challenge references form adaptation prototypes:
p'_c = normalize((1 - alpha) p_c + alpha r_c)
alpha=0.15 is selected only on validation. Thirteen of 21 classes use audio-trained prototypes; eight use text fallback. Sixteen AI pseudo-label candidates fail the fixed-validation gate, so production uses none.
Scope: 454 recordings from 13 classes with at least five source groups per class.
| Method | Accuracy | Macro-F1 | Top-three |
|---|---|---|---|
| BioLingual zero-shot | 88.11% | 83.94% | 96.26% |
| Prototype post-training | 94.05% | 91.95% | 99.12% |
| Class-balanced linear head | 94.71% | 93.09% | 99.34% |
Uploader-group bootstrap prototype-over-zero-shot gains:
Linear-head-versus-prototype intervals cross zero, so production retains prototypes.
Non-target animals are globally separated by uploader; environmental audio is separated by original ESC fold. Prototype-train, calibration, and test overlaps are zero.
| Group | AUROC | Online Known recall | Online FAR |
|---|---|---|---|
| Overall | 0.9696 | 72.73% | 6.45% |
| Non-target animal | 0.9500 | 72.73% | 20.00% |
| Environmental sound | 0.9789 | 72.73% | 0.00% |
Unknown is a conservative risk signal, not an expert species identification.