Scriba
Transcribe multilingual audio/video meetings locally into speaker-labeled Markdown with word times and voice clips. Uses fast Russian GigaAM or multilingual WhisperX, Nemotron diarization across languages with pyannote fallback, and optional validated host-AI punctuation. Rename speakers after identification. Use for transcripts, расшифровку, транскрипт, диаризацию, "кто что сказал", or транскрибировать/расшифровать встречу или запись.
Install
npx skills add AlexanderAbramovPav/scriba --skill scribagit clone --depth 1 https://github.com/AlexanderAbramovPav/scriba.git /tmp/scriba
mkdir -p ~/.claude/skills
cp -r /tmp/scriba/. ~/.claude/skills/scribaPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# scriba Turns any audio/video file into an accurate, speaker-diarized Markdown transcript. Two-pass: first produce `Speaker 1/2/3` + representative samples, then rename to real people. ## First-run setup — the AI handles it; the user only clicks 3 links and pastes a token **First check for local `nemo-speech` (or `SCRIBA_NEMO_BIN`).** Nemotron needs no HF token. If unavailable, or the user needs `--enroll` / `--speakers`, check whether `~/.config/scriba/hf_token` exists before selecting pyannote. If it does, skip this block. If it doesn't, the user needs a HuggingFace token *before* diarization can work — and they should NOT be left to figure that out alone. Walk them through it conversationally in their language (the chat's language is the user's language; default to English when ambiguous). The flow: > «Чтобы определить, кто что говорит, нужен бесплатный HuggingFace-токен (~30 секунд): > > 1. Открой <https://huggingface.co/join> — заведи аккаунт, если нет. > 2. Открой <https://huggingface.co/pyannote/segmentation-3.0> и нажми «Agree and access repository» (один клик). > 3. То же на <https://huggingface.co/pyannote/speaker-diarization-community-1> — это модель, которая распознаёт спикеров (один клик). > 4. Открой <https://hf.co/settings/tokens> → «+ Create new token» → имя `scriba`, тип **Read**, скопируй сгенерированный токен (вид `hf_...`). > 5. Вставь его сюда в чат — я сохраню локально с правами 600 и больше никогда не попрошу.»
Details
- Skill name
scriba- Repository
- AlexanderAbramovPav/scriba
- License
- MIT
- Repo updated
- 29 October 2026
FAQ
What does the Scriba skill do?
Transcribe multilingual audio/video meetings locally into speaker-labeled Markdown with word times and voice clips. Uses fast Russian GigaAM or multilingual WhisperX, Nemotron diarization across languages with pyannote fallback, and optional validated host-AI punctuation. Rename speakers after identification. Use for transcripts, расшифровку, транскрипт, диаризацию, "кто что сказал", or транскрибировать/расшифровать встречу или запись.
How do I install the Scriba skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add AlexanderAbramovPav/scriba --skill scriba. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Scriba popular?
It lives in AlexanderAbramovPav/scriba, which has 15 GitHub stars and 0 forks.
Related skills
Personal Ip Illustrations
14Jaimo-so/personal-ip-illustrations
为中文文章、帖子、博客、Notion 文档、工作流、方法论、流程、结构、状态、隐喻或观点设计并生成带固定个人 IP 主角的正文配图。用于用户提到“我的 IP 配图”“个人形象配图”“文章插图”“配图建议”“shot list”，或要求用随 Skill 提供的棕发圆框眼镜少年 IP 生成、编辑、去标题或改图；默认采用 16:9 纯白底、全手绘角色与结构、少量红橙蓝批注的清爽怪诞视觉语言。
Gwt
11allenk/gwt-integrations
Remove Gemini AI watermarks from images using GeminiWatermarkTool (GWT). Use this skill whenever the user mentions: Gemini watermark, 浮水印, AI image cleanup, removing logos from Gemini-generated images, Google ImageFX cleanup, batch processing AI-generated photos, or anything about a semi-transparent logo in the bottom-right corner of an image — even if they don't say "GeminiWatermarkTool" explicitly. Also trigger for follow-up requests like "the corner still looks dirty" or "there's still residual artifacts" after a previous removal attempt.
Storyframe Video Skill
14nexsjournal/storyframe-video
从 brief 到成片的标准视频生产系统，覆盖产品推广、知识讲解、口播优化、活动宣传、课程演示和音乐 MV。先接收并核验用户提供的文案、音乐、录屏、截图、Logo 与参考片，再用真实素材、故事方向、分镜、节拍表、声音设计和验收清单，产出 15–180 秒的 16:9 / 9:16 / 1:1 成片；包含 Suno 歌词制作单、Playwright + ffmpeg 确定性渲染和自动质检。当用户要求做宣传片、讲解片、活动片、课程演示、口播剪辑、产品片、MV，或要把一段文案/歌曲/录屏变成视频时使用。支持 UI 电影、动态字体、纸感手作、丝网印、像素世界等 12 种风格关键词。
Scroll Scrub Hero
14Black-Line-Ops/scroll-scrub-hero
Build a scroll-scrubbed hero from one or two REAL photos plus an idea — or from the idea alone. Scrolling drives a subject through a transformation (bare yard → finished pool, empty room → staged interior, raw stock → machined part, before → after). Uses kie.ai: GPT 5.6 Sol writes the storyboard, GPT Image 2 renders keyframe stills so the client's ACTUAL property stays recognisable (and draws the opening frame when there is no photo), Kling 3.0 first/last-frame tweens the motion, and ffmpeg cuts drop-in scrub frames plus config.js. Use whenever someone wants a scroll-driven build/progress/befo
Interview Transcript Organizer
12AiYuSherry/interview-transcript-organizer
将面试/访谈录音转写文本整理为规范 Word 文档，含面试复盘表。输出三版文档：面试总复盘（复盘表格）、问题总结（精简 Q&A）、完整记录（完整回答）。格式：中文楷体英文新罗马，问题蓝色加粗，回答黑色左对齐。自动按 `N-公司名面试 - YYYYMMDD/` 路径保存。
Motion Video Workflow
14isaachang/motion-video-workflow
用户提供配音（背景音乐可选，没有就自动合成），要做口播动效视频（案例分享、AI资讯、AI知识讲解、教程）时使用：听写、抓素材、分镜、代码动效，按需出16:9或9:16成片，并出3:4和4:3封面。
Browse the most popular Claude skills, all images, video & audio skills, or write your own with the skills generator.