ElevenLabs Voice Personas v2.1

Comprehensive voice synthesis toolkit using ElevenLabs API.

🚀 First Run - Setup Wizard

When you first use this skill (no config.json exists), run the interactive setup wizard:

CODEBLOCK0

The wizard will guide you through:

1. 1. API Key - Enter your ElevenLabs API key (required)
2. Default Voice - Choose from popular voices (Rachel, Adam, Bella, etc.)
3. Language - Set your preferred language (32 supported)
4. Audio Quality - Standard or high quality output
5. Cost Tracking - Enable usage and cost monitoring
6. Budget Limit - Optional monthly spending cap

🔒 Privacy: Your API key is stored locally in config.json only. It never leaves your machine and is automatically excluded from git via .gitignore.

To reconfigure at any time, simply run the setup wizard again.

✨ Features

• - 18 Voice Personas - Carefully curated voices for different use cases
• 32 Languages - Multi-language synthesis with the multilingual v2 model
• Streaming Mode - Real-time audio output as it generates
• Sound Effects (SFX) - AI-generated sound effects from text prompts
• Batch Processing - Process multiple texts in one go
• Cost Tracking - Monitor character usage and estimated costs
• Voice Design - Create custom voices from descriptions
• Pronunciation Dictionary - Custom word pronunciation rules
• OpenClaw Integration - Works with OpenClaw's built-in TTS

🎙 Available Voices

🎯 Quick Presets

• - default → rachel (warm, friendly)
• INLINECODE4 → adam (documentaries)
• INLINECODE5 → matilda (corporate)
• INLINECODE6 → george (audiobooks)
• INLINECODE7 → alice (tutorials)
• INLINECODE8 → brian (meditation)
• INLINECODE9 → liam (social media)
• INLINECODE10 → eric (business)
• INLINECODE11 → river (inclusive)
• INLINECODE12 → george
• INLINECODE13 → charlie
• INLINECODE14 → daniel (news)

🌍 Supported Languages (32)

The multilingual v2 model supports these languages:

CODEBLOCK1

💻 CLI Usage

Basic Text-to-Speech

CODEBLOCK2

Streaming Mode

Generate audio with real-time streaming (good for long texts):

CODEBLOCK3

Batch Processing

Process multiple texts from a file:

CODEBLOCK4

JSON batch format:
CODEBLOCK5

Simple text format (one per line):
CODEBLOCK6

Usage Statistics

CODEBLOCK7

🎵 Sound Effects (SFX)

Generate AI-powered sound effects from text descriptions:

CODEBLOCK8

Example prompts:

• - "Thunder rumbling in the distance"
• "Cat purring contentedly"
• "Typing on a mechanical keyboard"
• "Spaceship engine humming"
• "Coffee shop background chatter"

🎨 Voice Design

Create custom voices from text descriptions:

CODEBLOCK9

Voice Design Options:

📖 Pronunciation Dictionary

Customize how words are pronounced:

Edit pronunciations.json:
CODEBLOCK10

Usage:

# Pronunciations are applied automatically
python3 scripts/tts.py --text "The OpenClaw API is great" --voice rachel

# Disable pronunciations
python3 scripts/tts.py --text "The API is great" --voice rachel --no-pronunciations

💰 Cost Tracking

The skill tracks your character usage and estimates costs:

CODEBLOCK12

Output:

📊 ElevenLabs Usage Statistics

  Total Characters: 15,230
  Total Requests:   42
  Since:            2024-01-15

💰 Estimated Costs:
  Starter    $4.57 ($0.30/1k chars)
  Creator    $3.66 ($0.24/1k chars)
  Pro        $2.74 ($0.18/1k chars)
  Scale      $1.68 ($0.11/1k chars)

🤖 OpenClaw TTS Integration

Using with OpenClaw's Built-in TTS

OpenClaw has built-in TTS support that can use ElevenLabs. Configure in ~/.openclaw/openclaw.json:

CODEBLOCK14

Triggering TTS in Chat

In OpenClaw conversations:

• - Use /tts on to enable automatic TTS
• Use the tts tool directly for one-off speech
• Request "read this aloud" or "speak this"

Using Skill Scripts from OpenClaw

CODEBLOCK15

⚙ Configuration

The scripts look for API key in this order:

1. 1. ELEVEN_API_KEY or ELEVENLABS_API_KEY environment variable
2. Skill-local .env file (in the skill directory)

Create .env file:
CODEBLOCK16

Note: The skill no longer reads from ~/.openclaw/openclaw.json. Use environment variables or the skill-local .env file.

🎛 Voice Settings

Each voice has tuned settings for optimal output:

📝 Triggers

• - "use {voice_name} voice"
• "speak as {persona}"
• "list voices"
• "voice settings"
• "generate sound effect"
• "design a voice"

📁 Files

CODEBLOCK17

🔗 Links

• - ElevenLabs
• API Documentation
• Voice Library
• Sound Effects API
• Voice Design API

📋 Changelog

v2.1.0

• - Added interactive setup wizard (scripts/setup.py)
• Onboarding guides through API key, voice, language, quality, and budget settings
• Config stored locally in config.json (added to .gitignore)
• Professional, privacy-focused setup experience

v2.0.0

• - Added 32 language support with --lang parameter
• Added streaming mode with --stream flag
• Added sound effects generation (sfx.py)
• Added batch processing with --batch flag
• Added cost tracking with --stats flag
• Added voice design tool (voice-design.py)
• Added pronunciation dictionary support
• Added OpenClaw TTS integration documentation
• Improved error handling and progress output