htlin222

PDF To Tts Zh Skill — Development skill for Claude Code

Development community

Turn a PDF into a single narrated Traditional Chinese audiobook MP3 via edge-tts.

How to install PDF To Tts Zh Skill

This entry records only its repository, not the path inside it, so there is no exact command to give. Open htlin222/pdf-to-tts-zh-skill and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What PDF To Tts Zh Skill does

Turn a PDF into a single narrated Traditional Chinese audiobook MP3 via edge-tts. Install: npx skills add htlin222/pdf-to-tts-zh-skill

Alternatives in Development

Book To Skill — Turn any technical book PDF into a Claude Code skill — ready to study, reference, and use while you work 25.9k ★
Anything2explainer — Topic in, narrated explainer video out 151 ★
Iso 24495 Skill — ISO 24495-1 plain language rules as a Claude Code skill — bilingual: English + Traditional Chinese (繁體中文淺白改寫) 122 ★

README

pdf-to-tts-zh

[![Build & Release Skill](https://github.com/htlin222/pdf-to-tts-zh-skill/actions/workflows/release.yml/badge.svg)](https://github.com/htlin222/pdf-to-tts-zh-skill/actions/workflows/release.yml) [![GitHub Release](https://img.shields.io/github/v/release/htlin222/pdf-to-tts-zh-skill?include_prereleases&label=skill%20version)](https://github.com/htlin222/pdf-to-tts-zh-skill/releases/latest) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE) [![Skills Protocol](https://img.shields.io/badge/protocol-vercel--labs%2Fskills-blue)](https://github.com/vercel-labs/skills) [![Compatible Agents](https://img.shields.io/badge/agents-40%2B-green)](https://github.com/vercel-labs/skills#supported-agents)

Turn a PDF (slides, case reports, handouts) into a single narrated Traditional Chinese audiobook MP3 — page by page.

Install

npx skills add htlin222/pdf-to-tts-zh-skill
npx skills add -g htlin222/pdf-to-tts-zh-skill        # global
npx skills add htlin222/pdf-to-tts-zh-skill --agent claude-code  # specific agent

What it does

Given a PDF, the skill drives a four-stage pipeline:

Rasterize — pdftoppm renders each page to a PNG.
逐字稿 — the agent reads every page image and writes a natural spoken Traditional Chinese narration (one scripts/page-NN.txt per page). Technical tokens (drug names, abbreviations) stay in English for correct pronunciation; numbers and dates are verbalized in Chinese reading order. For long decks it fans out parallel subagents over page ranges with a shared narrator style.
TTS — edge-tts (free Microsoft Edge neural voices, no API key) synthesizes each page to an mp3; default voice zh-TW-HsiaoChenNeural, swappable.
Combine — ffmpeg concatenates the per-page mp3s into one continuous file.

Intermediate artifacts are kept so you can edit a single page's script and re-synthesize, or just re-stitch, without redoing the whole run. A `scripts/build.