yaya-diarize SKILL.md skillA
yaya-diarize from eatmoreduck/yaya-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add eatmoreduck/yaya-skills/yaya-diarize@v0.1.0Pins v0.1.0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- eatmoreduck/yaya-skills
- Path
skills/yaya-diarize/SKILL.md- Versions recorded
- 1
- Latest version
- v0.1.0, committed Oct 12, 2026
- Size
- 2,443 bytes
- Repository stars
- 9
- License
- MIT
Separates speakers in any audio entirely locally — outputs a who-spoke-when timeline (sherpa-onnx voiceprint clustering, no cloud, no transcript needed). Use when the user asks for "人声分离", "说话人分离", "谁在什么时候说话", "分轨", or wants speaker turns from an audio; also serves as the separation stage called by yaya-asr.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# yaya-diarize —— 说话人分离（谁在何时说话）
**音频进 → 说话人时间线出。** 只认声音、不认文字：sherpa-onnx（pyannote 分割 + CAM++ 中文声纹）聚类出"谁在什么时候说话"。纯本地，免费、零登录。
被 [yaya-asr](../yaya-asr/) 作为分离阶段调用（**与 MLX 转写并行跑**：GPU 转写 ∥ CPU 分离，互不争抢）；也可以单独使用。
## 何时使用
- 用户给一段录音，想知道"谁在什么时候说话"（不需要文字稿）
- 用户说"人声分离""说话人分离""分个轨""分一下说话人"
- 想给音频做说话人标注，文字用别的引擎转
**不适用**：要"谁说了什么"（文字+说话人）→ 直接用 `yaya-asr`（它内部自动并行调用本 skill）；要 Space 干货笔记 → `yaya-xspace`。
## 快速开始
```bash
bash {baseDir}/run.sh recording.m4a
bash {baseDir}/run.sh meeting.mp4 --speakers 2 -o ./out
```
依赖说明：分离引擎和模型住在 `yaya-asr` 的共享环境里，**首次使用前先跑一次 yaya-asr**（自动建 venv + 下模型 ~35MB，无需 HF token）。
## 产物
```
<stem>.turns.json {"num_speakers": N, "turns": [{"start", "end", "speaker"}]}
```
`speaker` 是按声纹聚类的编号（0..N-1），非真实姓名；不标注重叠语音（两人同时说归一人）。
## 参数
| 参数 | 说明 |
|---|---|
| `--speakers N` | **强制聚成 N 类**（引擎语义=确切人数，**不是上限**）。人少的录音务必填真实人数；不给默认 12 类；`-1` 按阈值自动聚类（长音频易过分割） |
| `-o DIR` | 输出目录（默认与输入同目录） |
| `--workdir DIR` | 缓存目录（默认 `.<stem>.diar`；yaya-asr 会传自己的缓存目录来共享） |
缓存带源文件指纹 + 聚类参数：换音频或换 `--speakers` 自动作废重算；坏缓存自愈。Read the whole file at its exact version.
How to install
mdr add eatmoreduck/yaya-skills/yaya-diarize@sha256:4af343b39cb22e7cThe line at the top of this page pins the label v0.1.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_tduhwszqd72zpg5u)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Oct 12, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (2443 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
eatmoreduck/yaya-skills · 9 stars · license MIT · pushed 2026-10-27 · branch master
API
GET https://markdownregistry.com/api/v1/artifacts/art_tduhwszqd72zpg5u GET https://markdownregistry.com/api/v1/resolve?ref=eatmoreduck/yaya-skills/yaya-diarize GET https://markdownregistry.com/api/v1/blob/4af343b39cb22e7cf105e19c467aad869d0f50efbf93d3264855bee53369b9be
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.