audiobook-production
について
このスキルは、ACX準拠のオーディオブック制作のためのツールとガイダンスを提供し、Audibleの技術基準を満たすことを保証します。開発者がプロフェッショナルなナレーションワークフローを構築し、魅力的な音声コンテンツを作るためのマルチメディア学習原理を適用することを支援します。配信用のオーディオブックを作成する際や、長文のテキストコンテンツをプロフェッショナルな音声に変換する際にご利用ください。
クイックインストール
Claude Code
推奨npx skills add guia-matthieu/clawfu-skills -a claude-code/plugin add https://github.com/guia-matthieu/clawfu-skillsgit clone https://github.com/guia-matthieu/clawfu-skills.git ~/.claude/skills/audiobook-productionこのコマンドをClaude Codeにコピー&ペーストしてスキルをインストールします
ドキュメント
Audiobook Production
Master ACX-compliant audiobook production using Audible's technical standards and Richard Mayer's multimedia learning principles for engaging, professional audio.
When to Use This Skill
- Producing an audiobook for Audible/ACX distribution
- Creating audio versions of courses or written content
- Setting up narration workflow for long-form content
- Ensuring audio meets professional distribution standards
- Self-producing an audiobook as an author
- Managing audiobook production projects
Methodology Foundation
Source: ACX (Audiobook Creation Exchange) + Richard Mayer (Multimedia Learning)
Core Principle: Professional audiobook production balances technical excellence (meeting strict ACX standards) with engaging delivery (applying learning science principles). As Mayer's research shows: "People learn better from friendly human voice than machine voice" (Voice Principle).
Why This Matters: ACX/Audible rejects audiobooks that don't meet technical specifications, wasting production time and money. Beyond technical compliance, understanding how listeners process audio enables production choices that maximize engagement and retention.
What Claude Does vs What You Decide
| Claude Does | You Decide |
|---|---|
| Structures production workflow | Final creative direction |
| Suggests technical approaches | Equipment and tool choices |
| Creates templates and checklists | Quality standards |
| Identifies best practices | Brand/voice decisions |
| Generates script outlines | Final script approval |
What This Skill Does
- Ensures ACX technical compliance - Specs for bit rate, sample rate, RMS, peaks, and noise floor
- Structures audiobook projects - Chapter organization, credits, retail samples
- Applies learning principles - Pacing, tone, and delivery that aid comprehension
- Manages production workflow - From script to final master
- Handles proofing and QC - Catching errors before submission
How to Use
Check ACX Compliance
Help me verify this audiobook meets ACX submission requirements.
[provide audio specs or describe setup]
Plan Audiobook Structure
Help me structure an audiobook from this manuscript:
Total word count: [X]
Chapters: [Y]
Type: [Fiction/Non-fiction]
Create Production Workflow
Create a production plan for my audiobook project:
Book length: [word count]
Narrator: [self/hired]
Deadline: [date]
Instructions
When producing audiobooks, follow this methodology:
Step 1: Understand ACX Technical Requirements
ACX has strict specifications that must be met for acceptance.
## ACX Audio Specifications (2026)
### File Format
- MP3, 192 kbps or higher (constant bit rate)
- 44.1 kHz sample rate
- Mono (single channel)
### Audio Levels
- RMS: -23dB to -18dB
- Peak values: -3dB maximum (no clipping)
- Noise floor: -60dB or lower
### File Naming
- Chapter files: [BookTitle]_[ChapterNumber].mp3
- Opening credits: [BookTitle]_Opening_Credits.mp3
- Closing credits: [BookTitle]_Closing_Credits.mp3
### Content Requirements
- Opening credits (first file): Title, author, narrator
- Closing credits (last file): "This has been [title] by [author], narrated by [narrator]"
- Each chapter file is standalone (no cross-file audio)
- Consistent room tone throughout
### Quality Requirements
- No extraneous sounds (mouth clicks, pops, page turns)
- Consistent audio quality across all files
- Retail sample: ~5 minutes of representative content
Critical: Run ACX Check tool on every file before submission.
Step 2: Prepare the Manuscript
Before recording, optimize the text for audio delivery.
## Manuscript Preparation Checklist
### Structure
□ Chapter breaks clearly marked
□ Section breaks identified (longer pause points)
□ Footnotes converted to in-text explanations or removed
□ Visual content (charts, tables) adapted for audio or noted for skip
### Pronunciation Guide
□ Character names with phonetic spelling
□ Foreign words and phrases
□ Technical terms and jargon
□ Place names and proper nouns
□ Acronyms expanded or spelled out
### Narrator Notes
□ Tone shifts marked
□ Character voice cues (if fiction)
□ Emphasis suggestions (minimal)
□ "As shown in Figure X" → adapted for audio
□ Author asides or commentary identified
### Timing Estimate
□ Word count per chapter
□ Total word count
□ Estimated runtime: words ÷ 9,300 = hours (approx)
- 50,000 words ≈ 5-6 hours
- 80,000 words ≈ 8-9 hours
Step 3: Set Up Recording Environment
Consistent, quality recording requires proper setup.
## Home Studio Essentials
### Minimum Equipment
- **Microphone**: USB condenser (AT2020, Blue Yeti) or XLR setup
- **Pop filter**: Essential for plosives
- **Mic stand**: Stable positioning
- **Headphones**: Closed-back, wired
### Room Treatment
- Record in smallest, most furnished room available
- Hang blankets on hard surfaces
- Use reflection filter behind mic
- Close windows, turn off HVAC
- Eliminate electronic hum (fridges, computers)
### Recording Software (DAW)
- Free: Audacity, GarageBand
- Professional: Adobe Audition, Reaper, Pro Tools
- ACX-specific: Hindenburg Pro (audiobook workflow)
### Test Recording Checklist
□ Record 30 seconds of room tone
□ Record test passage
□ Check RMS levels (-23 to -18 dB)
□ Check peak levels (below -3 dB)
□ Check noise floor (below -60 dB)
□ Listen for room echo or hum
□ Verify consistent mic positioning
Step 4: Apply Mayer's Multimedia Learning Principles
Science-backed techniques for engaging audio.
## Mayer's Principles for Audiobook Delivery
### Voice Principle
"People learn better from human voice than machine voice."
→ Warm, conversational delivery beats polished broadcast voice
→ Let personality come through
→ Avoid monotone "reading" voice
### Personalization Principle
"People learn better when words are conversational rather than formal."
→ Write/deliver as if talking to one person
→ Use "you" and "we"
→ Contractions are encouraged
### Segmenting Principle
"People learn better when content is in learner-paced segments."
→ Clear chapter breaks
→ Section pauses for complex material
→ Chapter summaries for dense non-fiction
### Redundancy Principle (for video/slides)
"People learn better from graphics and narration than graphics + narration + text."
→ For audiobooks: don't describe visual elements you can't show
→ Adapt visual content or acknowledge it's omitted
### Coherence Principle
"People learn better when extraneous material is excluded."
→ Cut verbal filler
→ Remove tangents that don't serve the chapter
→ Every word earns its place
Step 5: Production Workflow
Systematic process from script to master.
## Audiobook Production Pipeline
### Phase 1: Pre-Production (Week 1-2)
□ Manuscript finalized and proofread
□ Pronunciation guide complete
□ Chapter structure confirmed
□ Recording schedule created
□ Test recordings approved
### Phase 2: Recording (Weeks 2-6, varies by length)
□ Opening credits recorded first
□ Chapters recorded in order (usually)
□ Room tone captured each session
□ Daily backup of raw files
□ Notes logged for pickups needed
### Phase 3: Editing (Concurrent or following)
□ Remove mistakes and restart takes
□ Remove excessive breaths (keep natural ones)
□ Eliminate mouth clicks and pops
□ Smooth edits (crossfade)
□ Consistent pacing throughout
### Phase 4: Mastering
□ Noise reduction applied (light touch)
□ Compression for consistent levels
□ Limiting to prevent peaks
□ Final RMS check (-23 to -18 dB)
□ Export to ACX specs (192 kbps MP3, 44.1kHz, mono)
### Phase 5: Quality Control
□ Listen to every file start to finish
□ Check first/last 5 seconds of each file
□ Verify file naming convention
□ Run ACX Check or similar tool
□ Create retail sample (5 min)
Step 6: Credits and Sample
Specific requirements for opening, closing, and retail sample.
## Opening Credits Script
(Must be first audio file)
"[Book Title]"
Written by [Author Name]
Narrated by [Narrator Name]
[Optional: Publisher name]
[Optional: Copyright year]
---
## Closing Credits Script
(Must be last audio file)
"This has been [Book Title], written by [Author Name],
narrated by [Narrator Name]."
[Optional: "Visit [website] for more" or other approved callout]
[Optional: "Audiobook production by [producer]"]
---
## Retail Sample Guidelines
**Length**: Approximately 5 minutes
**Content**: Representative excerpt that hooks listeners
**Selection criteria**:
- Showcases narrator's ability
- Engages interest without major spoilers
- Demonstrates the book's voice/style
- Often from early chapters (context not required)
**Do NOT include**: Credits or chapter announcements in sample
Examples
Example 1: Self-Published Non-Fiction Audiobook
Context: An author is self-narrating their business book.
Input:
Help me plan audiobook production for my 55,000-word business book. I'm self-narrating. I have a Blue Yeti mic and record in my home office.
Output:
Production Plan: Business Book Audiobook
Project Specifications
- Word count: 55,000 words
- Estimated runtime: 5.5-6 hours
- Narrator: Author (self-narrated)
- Equipment: Blue Yeti USB mic (adequate for ACX)
Pre-Production (Days 1-3)
Manuscript Prep
- Review all 12 chapters for audio adaptation
- Convert/adapt 8 figures referenced in text
- Create pronunciation guide for industry terms
- Mark chapter breaks and section pauses
- Estimate 4,500 words/chapter average = ~29 min/chapter
Room Preparation
- Assess home office acoustics
- Add blanket on wall behind recording position
- Use closet with clothes as recording booth alternative
- Test recordings at 3 different times of day (HVAC, traffic noise)
Sample Script: "As shown in Figure 3" → "Here's how it works: imagine a funnel with three stages..."
Recording Schedule (Weeks 1-3)
Daily recording block: 2 hours max (voice fatigue) Target per session: 1.5-2 chapters (8,000 words) Total sessions needed: ~8-10
Schedule:
- Day 1: Opening credits + Chapter 1
- Day 2: Chapters 2-3
- Day 3: Chapters 4-5
- [continue pattern]
- Day 8: Chapter 12 + Closing credits
- Days 9-10: Pickups and problem sections
Session Ritual:
- Hydrate 30 min before (room temp water)
- Vocal warmup (5 min)
- Record room tone (10 sec)
- Test recording, check levels
- Record chapter(s)
- Log pickup notes
Post-Production (Week 4)
Editing workflow per chapter:
- Import raw audio
- Remove mistakes and alternate takes
- Clean up breaths and mouth clicks
- Apply light noise reduction
- Normalize RMS to -20 dB
- Check peaks (below -3 dB)
- Export as MP3 (192 kbps, 44.1 kHz, mono)
File naming:
- BusinessBookTitle_Opening_Credits.mp3
- BusinessBookTitle_001.mp3
- BusinessBookTitle_002.mp3
- ...
- BusinessBookTitle_012.mp3
- BusinessBookTitle_Closing_Credits.mp3
Quality Control (Days 1-2)
- Listen through every file with checklist
- Run ACX Check on all files
- Create 5-minute retail sample from Chapter 2
- Verify consistent sound across all chapters
- Final review of opening/closing credits
Technical Settings
Recording:
- 44.1 kHz, 24-bit WAV (edit in high quality)
- Mono input
- Gain set for peaks at -12 dB
Export (ACX compliant):
- MP3, 192 kbps CBR
- 44.1 kHz
- Mono
- RMS: -20 dB (within -23 to -18)
- Peak: -3 dB max
Example 2: ACX Troubleshooting
Context: Audiobook files are being rejected by ACX.
Input:
My audiobook submission was rejected. ACX says "RMS levels out of spec" and "noise floor too high." How do I fix this?
Output:
ACX Rejection Fix: RMS and Noise Floor
Diagnosis
RMS out of spec: Your audio is either too quiet (below -23 dB) or too loud (above -18 dB)
Noise floor too high: Background hiss or hum is above -60 dB
Step-by-Step Fix
Fix 1: RMS Levels
In Audacity:
- Select all audio (Ctrl+A)
- Effect → Loudness Normalization
- Set to: -20 dB RMS (middle of ACX range)
- Apply
In Adobe Audition:
- Select all
- Effects → Amplitude → Match Loudness
- Set Target Loudness to -20 LUFS
- Process
Fix 2: Noise Floor
If noise is mild (just room tone):
- Find 2-3 seconds of "silence" (room tone only)
- Select it as noise profile
- Apply noise reduction to full track
- Use minimal settings (6-12 dB reduction) to avoid artifacts
In Audacity:
- Effect → Noise Reduction → Get Noise Profile
- Select all → Noise Reduction → 6-12 dB, 6 Sensitivity
If noise is significant:
- Re-record in quieter environment
- Use noise gate (removes sound below threshold)
- Consider iZotope RX for professional repair
Fix 3: Verify Peaks
After RMS adjustment, peaks may exceed -3 dB:
- Effect → Limiter
- Set ceiling to -3.0 dB
- Apply
Verification Workflow
- Make fixes to one chapter first
- Run ACX Check plugin or online tool
- Verify all specs pass:
- RMS: -23 to -18 dB ✓
- Peak: -3 dB or lower ✓
- Noise floor: -60 dB or lower ✓
- Apply same process to all files
- Re-check each file before resubmitting
Prevention for Future
Recording checklist:
- Noise floor below -60 dB before recording
- Gain set so peaks hit -12 dB when speaking
- Room tone recorded each session
- Test file checked before full session
Checklists & Templates
ACX Pre-Submission Checklist
## Technical Compliance
□ All files are 192 kbps MP3, 44.1 kHz, mono
□ RMS levels between -23 and -18 dB
□ Peak values at or below -3 dB
□ Noise floor at or below -60 dB
□ No clipping or distortion
□ Consistent levels across all chapters
## File Structure
□ Opening credits file present
□ Closing credits file present
□ Files named correctly: [Title]_[Number].mp3
□ Chapter numbers are sequential
□ No missing chapters
□ Retail sample created (≈5 minutes)
## Content Quality
□ First/last 5 seconds of each file are clean (room tone or fade)
□ No extraneous sounds (clicks, pops, breaths)
□ No room echo or reverb issues
□ Consistent voice quality throughout
□ All pickups and corrections completed
□ Credits read correctly (title, author, narrator)
## Final Review
□ Listened to complete audiobook
□ Spot-checked transitions between chapters
□ Verified retail sample is representative
□ Cover art meets ACX specs (if applicable)
□ All metadata entered correctly
Production Schedule Template
## Audiobook Production Schedule
**Title**: ________________________________
**Word count**: ___________ | **Est. runtime**: ___________
**Narrator**: _____________________________
**Target completion**: ____________________
### Pre-Production: Week ___
| Task | Due | Done |
|------|-----|------|
| Manuscript finalized | | □ |
| Pronunciation guide | | □ |
| Room/equipment tested | | □ |
| Opening credits scripted | | □ |
### Recording: Weeks ___ - ___
| Session | Chapters | Words | Est. Time | Done |
|---------|----------|-------|-----------|------|
| 1 | Credits + Ch 1 | | | □ |
| 2 | Ch 2-3 | | | □ |
| 3 | Ch 4-5 | | | □ |
| ... | | | | □ |
### Post-Production: Week ___
| Task | Due | Done |
|------|-----|------|
| Editing complete | | □ |
| Mastering complete | | □ |
| ACX specs verified | | □ |
| QC listen-through | | □ |
| Retail sample created | | □ |
### Submission: ___________
Skill Boundaries
What This Skill Does Well
- Structuring audio production workflows
- Providing technical guidance
- Creating quality checklists
- Suggesting creative approaches
What This Skill Cannot Do
- Replace audio engineering expertise
- Make subjective creative decisions
- Access or edit audio files directly
- Guarantee commercial success
References
- ACX. "Audio Submission Requirements" - Official technical specifications
- Richard Mayer. "Multimedia Learning" (2009) - Scientific principles for audio learning
- iZotope. "Professional Voice Over Recording Tips" - Technical recording guidance
- Audacity Team. "Audacity Manual" - Free editing software
Related Skills
- voiceover-direction - Directing talent for quality delivery
- audio-editing - Post-production fundamentals
- voice-design - AI voice alternatives for narration
- transcription-to-content - Repurposing audiobook content
Skill Metadata (Internal Use)
name: audiobook-production
category: audio
subcategory: voiceover
version: 1.0
author: MKTG Skills
source_expert: ACX, Richard Mayer
source_work: ACX Submission Requirements, Multimedia Learning
difficulty: intermediate
estimated_value: $500-2,000 per audiobook (equivalent production management)
tags: [audiobook, acx, audible, narration, production]
created: 2026-01-26
updated: 2026-01-26
GitHub リポジトリ
関連スキル
content-collections
メタこのスキルは、Content Collections(Markdown/MDXファイルを型安全なデータコレクションに変換するTypeScriptファーストのツール)の本番環境でテストされた設定を提供します。Zodバリデーションによる型安全性を実現し、ブログ、ドキュメントサイト、コンテンツ重視のVite + Reactアプリケーション構築時にご利用ください。Viteプラグインの設定、MDXコンパイルから、デプロイ最適化、スキーマバリデーションまで、すべてを網羅しています。
polymarket
メタこのスキルは、開発者がPolymarket予測市場プラットフォームを活用したアプリケーション構築を可能にします。API統合による取引や市場データの取得に加え、WebSocketを介したリアルタイムデータストリーミングにより、ライブ取引や市場活動を監視できます。取引戦略の実装や、ライブ市場更新を処理するツールの作成にご利用ください。
creating-opencode-plugins
メタこのスキルは、開発者がコマンド、ファイル、LSP操作など25種類以上のイベントタイプにフックするOpenCodeプラグインを作成することを支援します。JavaScript/TypeScriptモジュール向けに、プラグイン構造、イベントAPI仕様、および実装パターンを提供します。カスタムイベント駆動ロジックでOpenCode AIアシスタントのライフサイクルをインターセプト、監視、または拡張する必要がある場合にご利用ください。
sglang
メタSGLangは、高性能なLLMサービングフレームワークであり、RadixAttentionプレフィックスキャッシュを活用したJSON、正規表現、エージェントワークフロー向けの高速で構造化された生成を特長とします。特にプレフィックスが繰り返されるタスクにおいて、大幅に高速な推論を実現し、複雑な構造化出力やマルチターン対話に最適です。制約付きデコードが必要な場合や、広範なプレフィックス共有を伴うアプリケーションを構築する場合は、vLLMなどの代替案ではなくSGLangを選択してください。
