What is tts-synthesis?

A tool that turns course transcripts into spoken audio files. TTS, or text-to-speech, converts written text into speech; the workflow can split audio by slide and join it into one full recording.

Use it to create per-slide MP3 files and a complete MP3 from a transcript containing slide and pause markers.

It avoids manually recording or cutting narration and preserves pauses and slide boundaries from the transcript.

How tobyilee/course-builder configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to tobyilee/course-builder. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/tobyilee/course-builder/main/.claude/skills/tts-synthesis/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/tobyilee/course-builder

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/tobyilee/course-builder/tts-synthesis.svg)](https://agentmods.dev/skills/tobyilee/course-builder/tts-synthesis)

Your own site

<a href="https://agentmods.dev/skills/tobyilee/course-builder/tts-synthesis"><img src="https://agentmods.dev/badge/skills/tobyilee/course-builder/tts-synthesis.svg" alt="Measured on agentmods" height="20"></a>

Per session 128 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,326 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00128	$0.01326
Opus 5	$0.00064	$0.00663
Sonnet 5	$0.00026	$0.00265
Haiku 4.5	$0.00013	$0.00133

Measured 6d ago against content hash 2c9333b7609d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

tts-synthesis 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 6d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/run.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

.claude/skills/tts-synthesis/SKILL.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TTS Synthesis — Pluggable Audio Pipeline

transcript.txt를 per-slide MP3 + full.mp3로 합성한다. OpenAI gpt-4o-mini-tts + nova + speed 1.3가 기본. edge-tts는 offline/무료 대안.

왜 이 스킬이 존재하는가

원래 하네스의 Non-Goal이었으나 Phase 7에서 확장. TTS 합성은 (1) transcript 마커 파싱, (2) 엔진별 API 호출, (3) pause 무음 삽입 + concat, (4) retry가 모두 얽혀있어 스크립트화하지 않으면 재사용이 어렵다.

실행

표준 호출

python3 scripts/synthesize-tts.py \
  course/sections/<sec>/classes/<cls>/transcript.txt \
  course/sections/<sec>/classes/<cls>/audio

위는 기본값 (openai + nova + speed 1.3 + default 친근 ko instructions)을 사용.

래퍼 (pipefail 보장)

bash .claude/skills/tts-synthesis/scripts/run.sh \
  <transcript.txt> <out_dir> [추가 플래그]

래퍼가 set -o pipefail + tee의 실제 exit 전파 + .env 로딩을 책임진다.

엔진 전환

# edge-tts로 (오프라인, rate limit 있음)
--engine edge --voice ko-KR-SunHiNeural

# OpenAI 다른 voice/모델
--voice shimmer
--model tts-1-hd   # 고음질, 비용 2배
--speed 1.0        # 원속도

입력 마커 규격

[slide N] (줄 단독, N=1..) — TTS 후처리용 오디오 구간 분할 기준
[pause:Xms] — 무음 삽입 (기본 300~800). speed 파라미터에 영향받지 않음 (원본 유지, 사용자 확인된 선호)
[emph]...[/emph] — SSML 변환 시 <emphasis> (OpenAI는 instructions로 대체)

출력 구조

<out_dir>/
  slide_01.mp3  ~  slide_NN.mp3   # 슬라이드별
  full.mp3                          # 전체 concat

에이전트/오케스트레이터 연결

이 스킬은 tts-synthesizer 에이전트의 단일 책임 스킬. course-builder 오케스트레이터 Phase 5의 step 4 로 통합되어 있으며, asset-builder 가 build-bundle.sh 안에서 자동 위임 호출한다:

Phase 5 (build):
  step 1 gate (asset-builder)
  step 2 Marp HTML (asset-builder)
  step 3 Marp PNG (asset-builder, SKIP_PLAYER 가능)
  step 4 TTS 합성 ← tts-synthesizer (이 스킬), per-class
  step 5 manifest (asset-builder)
  step 6 player HTML (asset-builder, SKIP_PLAYER 가능)
  step 7 SSML 검증 (asset-builder)
  step 8 bundle.zip (asset-builder)

자동 호출 조건 (모두 충족):

coherence_report.overall == "pass" — gate 통과
OPENAI_API_KEY 존재 (없으면 edge-tts 폴백)
SKIP_TTS != 1
audio/full.mp3 미존재 또는 FORCE_TTS=1

standalone 호출(빌드 외부 사용자 트리거)도 지원 — tts-synthesizer 에이전트를 직접 invoke.

Read the full file on GitHub · 100 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

scripts/run.sh 898 B runs code

Changes

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.

6d ago First seen · 100 lines · 128 tokens per session scan A 2c9333b7609d

Subscribe to this mod's changes

tts-synthesis is a skill published in the GitHub repository tobyilee/course-builder (22 stars, last pushed 4mo ago), licensed MIT. It adds 128 tokens to every session and 1,326 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 2026-08-30.

Creative and media Learning and teaching

baoyu-comic

Knowledge comics (知识漫画): educational, biography, tutorial.

NousResearch/hermes-agent · 17 tokens

html-ppt-zhangzara-cartesian

An economics senior thesis on the employment effects of local minimum-wage increases — identification strategy, evidence, and limitations. Built as a decision-grade coursework defense deck for thesis committee.

nexu-io/open-design · 46 tokens

html-ppt-zhangzara-daisy-days

A customer-success workshop onboarding users to a project-management app — the first-value path and the habits that retain. Built as a decision-grade professional training deck for new customers, CS team.

nexu-io/open-design · 49 tokens

html-ppt-zhangzara-playful

A retail sales-floor training on consultative selling — the flow, the role-plays, and the daily habit that lifts conversion. Built as a decision-grade professional training deck for store associates, floor managers.

nexu-io/open-design · 52 tokens

html-ppt-zhangzara-retro-windows

An IT security-awareness training on spotting phishing — the tells, the drill, and what to do in the first 60 seconds. Built as a decision-grade professional training deck for all employees.

nexu-io/open-design · 51 tokens

sprite-animation

A pixel / sprite-style animated explainer slide — full-bleed cream stage, bold display year, animated pixel-art mascot (e.g. Hanafuda card, mushroom, or 8-bit console), kinetic Japanese display type, ticking timeline ribbon. Reads like a single frame of an educational motion video — looping CSS keyframes, no JS, ready…

nexu-io/open-design · 141 tokens

course-builder: Skill for Claude Code