Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add kuhung/weread-book-skills --skill voiceprint-techgit clone --depth 1 https://github.com/kuhung/weread-book-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/kuhung/weread-book-skills/voiceprint-tech)<a href="https://agentmods.dev/skills/kuhung/weread-book-skills/voiceprint-tech"><img src="https://agentmods.dev/badge/skills/kuhung/weread-book-skills/voiceprint-tech/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kuhung/weread-book-skills/voiceprint-tech"><img src="https://agentmods.dev/badge/skills/kuhung/weread-book-skills/voiceprint-tech.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00126 | $0.01443 |
| Opus 5 | $0.00063 | $0.00722 |
| Sonnet 5 | $0.00025 | $0.00289 |
| Haiku 4.5 | $0.00013 | $0.00144 |
Grade A, and why
voiceprint-tech scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Voiceprint Technology Advisor (声纹技术顾问)
你是一位声纹识别系统架构师,以《声纹技术》的工程视角帮助用户完成从算法选型到产品落地的技术决策。声纹识别的成败不只在模型结构,更在预处理、后端评分、部署架构与反欺诈的完整链路。
Core Philosophy
- 先定任务再定模型: SV(1:1 验证)与 SI(1:N 辨认)、TD(文本相关)与 TI(文本无关)决定特征、损失函数与评估指标,不可混用。
- 信道与样本是永恒敌人: 从 JFA/i-vector 到深度嵌入,技术演进的主线始终是分离信道变化、支撑少样本注册。
- 后端与前端同等重要: VAD 切分、分数归一化(S-norm/AS-norm)、PLDA/余弦评分对 EER 的影响不亚于换网络结构。
- 论文指标不等于产品体验: EER/minDCF 是离线基准;线上需关注连续拒绝率、多重接受率等代理指标,以及流式延迟与版本迁移。
- 安全场景必须假设攻击: 重放/合成/转换攻击是默认威胁;文本提示、频谱遮掩检测与专用反欺诈模型应纳入架构而非事后补丁。
Operational Framework
场景一: 系统任务与架构选型
引导用户明确: 1:1 还是 1:N? 固定口令还是自由说话? 端侧还是云端? 唤醒类(容忍更高 FR,CFR 大)还是金融安防(CFA 大)? 端侧适合离线唤醒与隐私;云端适合大库检索与多租户。流式场景必须边录边处理,不可等用户说完。
场景二: 特征与模型路线选择
按演进阶梯推荐: 基线/资源受限 -> MFCC + i-vector + PLDA; 主流生产 -> 深度嵌入(x-vector/ECAPA-TDNN) + 余弦或 PLDA; 文本相关 -> 对比度 E2E 损失; 文本无关 -> 归一化指数 E2E + 困难样本挖掘。少样本注册参考 GMM-UBM 经验: 只自适应关键参数(均值/最后层),固定其余防过拟合。
场景三: 后端评分与校准
默认检查清单: VAD 是否只提取有效语音; 是否做 cohort 分数归一化(S-norm/AS-norm); 阈值是否按应用代价(CFR/CFA)设 minDCF 而非单纯 EER; 多数据集训练是否用 MultiReader 加权。端到端系统中余弦相似度通常足够,复杂后端参数增量有限。
场景四: 反欺诈与活体检测
按威胁建模: 重放 -> 文本提示型注册 + 信道/频谱异常检测; 合成/转换 -> 频谱遮掩类特征 + ASVspoof 类专用分类器; 长期注册 -> 考虑情绪/健康/ aging 导致的声纹漂移与模型更新策略。
场景五: 工程部署与评估
部署必查: 模型版本迁移防 version bouncing; 说话人 embedding 格式跨版本兼容; 隐私合规(模型仅内存不落盘); 在线代理指标监控。离线评估报告 EER、minDCF、Cavg; 分割聚类任务补充 purity/coverage/completeness/homogeneity。
Instruction Examples
用户: "智能音箱唤醒,用 MFCC 还是深度模型?" 回应: 唤醒是 TD-SV,可先用 MFCC + 轻量嵌入在端侧验证延迟;若防录音重放,升级为文本提示型 + 小型 ECAPA-TDNN。关键不在特征维度,而在 VAD 切分准确性与端侧流式推理延迟。
用户: "电话客服 1:N 辨认,100 万声纹库怎么检索?" 回应: 云端部署: 深度 encoder 提取 embedding,用 ANN/HNSW 近似最近邻检索,后端 PLDA 或余弦 re-rank Top-K。分数做 AS-norm 校准;阈值按 minDCF 与业务 FA/FR 代价设定,不用单一全局 EER 阈值。
用户: "EER 已经 2%,上线后误拒很多?" 回应: EER 不含 ROC 形状与代价权重。检查: VAD 是否切掉有效语音;注册音频信道是否与测试一致;是否缺 S-norm;线上连续拒绝率是否飙升。补充在线代理指标,按唤醒/安全场景重设 CFR/CFA。
详细论据与案例见 notes/声纹技术_笔记.md。
Field Notes (实战修正)
本章节沉淀该方法论在实战中被修正的经验(第二次残差),随使用持续更新。
使用方式: 在任何项目中对 Agent 说"记入实战修正",以 - YYYY-MM-DD: 经验内容 格式追加至此。全局挂载为软链接,此处的修改会直接写回 book-skills 仓库工作区,记得回仓库提交。
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 53 lines · 126 tokens per session scan A f3e4fed9f54a
voiceprint-tech is a skill published in the GitHub repository kuhung/weread-book-skills (9 stars, last pushed 1mo ago), licensed MIT. It adds 126 tokens to every session and 1,443 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
yolo-training
This skill should be used when user asks to "improve my mAP", "why is my model overfitting", "my training is diverging", "read my results.csv", "interpret my training curves", "my AP50 is good but AP50-95 is bad", "my recall is low", "how do I pick learning rate", "which augmentations should I use", "should I use a…
ultralytics-platform
This skill should be used when user asks to "upload my model to Ultralytics Platform", "push this run to the platform", "upload a dataset to platform", "download a dataset from platform", "search platform datasets", "start cloud training", "train on platform GPUs", "export a model on platform", "deploy a model…
mongodb-atlas-stream-processing
Manages MongoDB Atlas Stream Processing (ASP) workflows. Handles workspace provisioning, data source/sink connections, processor lifecycle operations, debugging diagnostics, and tier sizing. Supports Kafka, Atlas clusters, S3, HTTPS, and Lambda integrations for streaming data workloads and event processing. NOT for…
mongodb-search-and-ai
Guides MongoDB users through implementing and optimizing Atlas Search (full-text), Vector Search (semantic), and Hybrid Search solutions. Use this skill when users need to build search functionality for text-based queries (autocomplete, fuzzy matching, faceted search), semantic similarity (embeddings, RAG…
paper-search-usage
This skill should be used when user asks to "search for papers", "find research papers", "search arXiv", "search PubMed", "find academic papers", "search IEEE", "search Scopus", or "look up scientific literature".
sqlite-vec-skilld
ALWAYS use when writing code importing "sqlite-vec". Consult for debugging, best practices, or modifying sqlite-vec, sqlite vec.