clawvox – OpenClaw Skill

---

name: clawvox description: ClawVox - ElevenLabs voice studio for OpenClaw. Generate speech, transcribe audio, clone voices, create sound effects, and more. homepage: https://elevenlabs.io/developers metadata: { "openclaw": { "emoji": "🎙️", "skillKey": "clawvox", "requires": { "bins": ["curl", "jq"], "env": ["ELEVENLABS_API_KEY"] }, "primaryEnv": "ELEVENLABS_API_KEY" } }

ClawVox

Transform your OpenClaw assistant into a professional voice production studio with ClawVox - powered by ElevenLabs.

Quick Reference

Action	Command	Description
Speak	{baseDir}/scripts/speak.sh 'text'	Convert text to speech
Transcribe	{baseDir}/scripts/transcribe.sh audio.mp3	Speech to text
Clone	{baseDir}/scripts/clone.sh --name "Voice" sample.mp3	Clone a voice
SFX	{baseDir}/scripts/sfx.sh "thunder storm"	Generate sound effects
Voices	{baseDir}/scripts/voices.sh list	List available voices
Dub	{baseDir}/scripts/dub.sh --target es audio.mp3	Translate audio
Isolate	{baseDir}/scripts/isolate.sh audio.mp3	Remove background noise

Setup

1. Get your API key from elevenlabs.io/app/settings/api-keys
2. Configure in ~/.openclaw/openclaw.json:

{
  skills: {
    entries: {
      "clawvox": {
        apiKey: "YOUR_ELEVENLABS_API_KEY",
        config: {
          defaultVoice: "Rachel",
          defaultModel: "eleven_turbo_v2_5",
          outputDir: "~/.openclaw/audio"
        }
      }
    }
  }
}

Or set the environment variable:

export ELEVENLABS_API_KEY="your_api_key_here"

Voice Generation (TTS)

Basic Text-to-Speech

# Quick speak with default voice (Rachel)
{baseDir}/scripts/speak.sh 'Hello, I am your personal AI assistant.'

# Specify voice by name
{baseDir}/scripts/speak.sh --voice Adam 'Hello from Adam'

# Save to file
{baseDir}/scripts/speak.sh --out ~/audio/greeting.mp3 'Welcome to the show'

# Use specific model
{baseDir}/scripts/speak.sh --model eleven_multilingual_v2 'Bonjour'

# Adjust voice settings
{baseDir}/scripts/speak.sh --stability 0.5 --similarity 0.8 'Expressive speech'

# Adjust speed
{baseDir}/scripts/speak.sh --speed 1.2 'Faster speech'

# Use multilingual model for other languages
{baseDir}/scripts/speak.sh --model eleven_multilingual_v2 --voice Rachel 'Hola, que tal'
{baseDir}/scripts/speak.sh --model eleven_multilingual_v2 --voice Adam 'Guten Tag'

Voice Models

Model	Latency	Languages	Best For
eleven_flash_v2_5	~75ms	32	Real-time, streaming
eleven_turbo_v2_5	~250ms	32	Balanced quality/speed
eleven_multilingual_v2	~500ms	29	Long-form, highest quality

Available Voices

Premade voices: Rachel, Adam, Antoni, Bella, Domi, Elli, Josh, Sam, Callum, Charlie, George, Liam, Matilda, Alice, Bill, Brian, Chris, Daniel, Eric, Jessica, Laura, Lily, River, Roger, Sarah, Will

Long-Form Content

# Generate audio from text file
{baseDir}/scripts/speak.sh --input chapter.txt --voice "George" --out audiobook.mp3

Speech-to-Text (Transcription)

Basic Transcription

# Transcribe audio file
{baseDir}/scripts/transcribe.sh recording.mp3

# Save to file
{baseDir}/scripts/transcribe.sh --out transcript.txt audio.mp3

# Transcribe with language hint
{baseDir}/scripts/transcribe.sh --language es spanish_audio.mp3

# Include timestamps
{baseDir}/scripts/transcribe.sh --timestamps podcast.mp3

Supported Formats

• MP3, MP4, MPEG, MPGA, M4A, WAV, WebM
• Maximum file size: 100MB

Voice Cloning

Instant Voice Clone

# Clone from single sample (minimum 30 seconds recommended)
{baseDir}/scripts/clone.sh --name MyVoice recording.mp3

# Clone with description
{baseDir}/scripts/clone.sh --name BusinessVoice \
  --description 'Professional male voice' \
  sample.mp3

# Clone with labels
{baseDir}/scripts/clone.sh --name MyVoice \
  --labels '{"gender":"male","age":"adult"}' \
  sample.mp3

# Remove background noise during cloning
{baseDir}/scripts/clone.sh --name CleanVoice \
  --remove-bg-noise \
  sample.mp3

# Test cloned voice
{baseDir}/scripts/speak.sh --voice MyVoice 'Testing my cloned voice'

Voice Library Management

# List all available voices
{baseDir}/scripts/voices.sh list

# Get voice details
{baseDir}/scripts/voices.sh info --name Rachel
{baseDir}/scripts/voices.sh info --id 21m00Tcm4TlvDq8ikWAM

# Search voices (filter output with grep)
{baseDir}/scripts/voices.sh list | grep -i "female"

# Filter by category
{baseDir}/scripts/voices.sh list --category premade
{baseDir}/scripts/voices.sh list --category cloned

# Download voice preview
{baseDir}/scripts/voices.sh preview --name Rachel -o preview.mp3

# Delete custom voice
{baseDir}/scripts/voices.sh delete --id "voice_id"

Sound Effects

# Generate sound effect
{baseDir}/scripts/sfx.sh 'Heavy rain on a tin roof'

# With duration
{baseDir}/scripts/sfx.sh --duration 5 'Forest ambiance with birds'

# With prompt influence (higher = more accurate)
{baseDir}/scripts/sfx.sh --influence 0.8 'Sci-fi laser gun firing'

# Save to file
{baseDir}/scripts/sfx.sh --out effects/thunder.mp3 'Rolling thunder'

Note: Duration range is 0.5 to 22 seconds (rounded to nearest 0.5)

Voice Isolation

# Remove background noise and isolate voice
{baseDir}/scripts/isolate.sh noisy_recording.mp3

# Save to specific file
{baseDir}/scripts/isolate.sh --out clean_voice.mp3 meeting_recording.mp3

# Don't tag audio events
{baseDir}/scripts/isolate.sh --no-audio-events recording.mp3

Requirements:

• Minimum duration: 4.6 seconds
• Supported formats: MP3, WAV, M4A, OGG, FLAC

Dubbing (Multi-Language Translation)

# Dub audio to Spanish
{baseDir}/scripts/dub.sh --target es audio.mp3

# Dub with source language specified
{baseDir}/scripts/dub.sh --source en --target ja video.mp4

# Check dubbing status
{baseDir}/scripts/dub.sh --status --id "dubbing_id"

# Download dubbed audio
{baseDir}/scripts/dub.sh --download --id "dubbing_id" --out dubbed.mp3

Supported languages: en, es, fr, de, it, pt, pl, hi, ar, zh, ja, ko, nl, ru, tr, vi, sv, da, fi, cs, el, he, id, ms, no, ro, uk, hu, th

API Usage Examples

For direct API access, all scripts use curl under the hood:

# Direct TTS API call
curl -X POST "https://api.elevenlabs.io/v1/text-to-speech/VOICE_ID" \
  -H "xi-api-key: $ELEVENLABS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "Hello world", "model_id": "eleven_turbo_v2_5"}' \
  --output speech.mp3

Error Handling

All scripts provide helpful error messages:

• 401: Authentication failed - Check your API key
• 403: Permission denied - Your API key may not have access
• 429: Rate limit exceeded - Wait before trying again
• 500/502/503: ElevenLabs API issues - Try again later

Testing

Run the test suite to verify everything works:

{baseDir}/test.sh YOUR_API_KEY

Or with environment variable:

export ELEVENLABS_API_KEY="your_key"
{baseDir}/test.sh

Troubleshooting

Common Issues

1. "exec host not allowed (requested gateway)"

   • The skill needs to run commands in a sandbox environment
   • Configure OpenClaw to use sandbox: tools.exec.host: "sandbox"
   • Or enable sandboxing in your OpenClaw config
   • Alternative: Configure exec approvals for gateway host (see OpenClaw docs)

2. Parse errors with quotes or exclamation marks

   • Use single quotes instead of double quotes: 'Hello world' not "Hello world!"
   • Avoid exclamation marks (!) in text when using double quotes
   • For complex text, use the --input option with a file

3. "ELEVENLABS_API_KEY not set"

   • Ensure ELEVENLABS_API_KEY is set or configured in openclaw.json
   • Check that the API key is at least 20 characters long

4. "jq is required but not installed"

   • Install jq: apt-get install jq (Linux) or brew install jq (macOS)

5. "Rate limited"

   • Check your ElevenLabs plan quota at elevenlabs.io/app/usage
   • Free tier: ~10,000 characters/month

6. "Voice not found"

   • Use {baseDir}/scripts/voices.sh list to see available voices
   • Check if the voice ID is correct

7. "Dubbing failed"

   • Ensure source audio is clear and audible
   • Check supported language codes

8. "File too large"

   • Transcription: 100MB max
   • Dubbing: 500MB max
   • Voice cloning: 50MB per file

Debug Mode

# Enable verbose output
DEBUG=1 {baseDir}/scripts/speak.sh 'test'

# Show API request details
DEBUG=1 {baseDir}/scripts/transcribe.sh audio.mp3

Pricing Notes

ElevenLabs API pricing (approximate):

• Flash v2.5: ~$0.06/min
• Turbo v2.5: ~$0.06/min
• Multilingual v2: ~$0.12/min
• Voice cloning: Included in plan
• Sound effects: ~$0.02/generation
• Transcription: ~$0.02/min (Scribe v1)

Free tier: ~10,000 characters/month

Links

• ElevenLabs Dashboard
• API Documentation
• Voice Library
• Pricing

ElevenLabs Voice Studio for OpenClaw

A comprehensive voice production studio skill for OpenClaw, powered by ElevenLabs. Transform your personal AI assistant into a full-featured voice platform with text-to-speech, speech-to-text, voice cloning, sound effects, and multilingual dubbing capabilities.

🎙️ Features

• Text-to-Speech (TTS) - Generate lifelike speech with multiple voice models
• Speech-to-Text (STT) - Transcribe audio with high accuracy
• Voice Cloning - Clone voices from audio samples
• Sound Effects - Generate custom audio effects from text descriptions
• Voice Isolation - Remove background noise from audio
• Multilingual Dubbing - Translate and dub audio/video to 32+ languages
• Voice Library Management - Browse and manage available voices

🚀 Quick Start

Installation

1. Copy this skill to your OpenClaw skills directory:

cp -r elevenlabs-voice-studio ~/.openclaw/skills/

2. Set your ElevenLabs API key:

export ELEVENLABS_API_KEY="your_api_key_here"

Or configure in ~/.openclaw/openclaw.json:

{
  skills: {
    entries: {
      "elevenlabs-voice-studio": {
        apiKey: "your_api_key_here"
      }
    }
  }
}

3. Get your API key from elevenlabs.io/app/settings/api-keys

Basic Usage

# Text to Speech
elevenlabs speak "Hello from ElevenLabs Voice Studio!"
elevenlabs speak -v Adam "Hello from Adam's voice"

# Transcribe Audio
elevenlabs transcribe meeting.mp3
elevenlabs transcribe -o transcript.txt recording.mp3

# Clone a Voice
elevenlabs clone -n MyVoice sample.mp3

# List Voices
elevenlabs voices list
elevenlabs voices info --name Rachel

# Generate Sound Effects
elevenlabs sfx "Thunder storm with heavy rain"

# Remove Background Noise
elevenlabs isolate noisy_recording.mp3 -o clean.mp3

# Dub to Another Language
elevenlabs dub -t es audio.mp3  # Spanish
elevenlabs dub -s en -t ja video.mp4  # English to Japanese

📚 Available Commands

speak - Text to Speech

elevenlabs speak [options] <text>

Options:
  -v, --voice <name>        Voice name or ID (default: Rachel)
  -m, --model <model>       TTS model: eleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2
  -o, --out <file>          Output file path
  -i, --input <file>        Read text from file
  --stability <0-1>         Voice stability
  --similarity <0-1>        Similarity boost
  --style <0-1>             Style exaggeration
  --speaker-boost           Enable speaker boost

transcribe - Speech to Text

elevenlabs transcribe [options] <audio_file>

Options:
  -o, --out <file>          Output file path
  -l, --language <code>     Language hint (en, es, fr, etc.)
  -t, --timestamps          Include word timestamps

clone - Voice Cloning

elevenlabs clone [options] <sample_files...>

Options:
  -n, --name <name>         Name for the cloned voice (required)
  -d, --description <text>  Voice description
  -l, --labels <json>       Labels as JSON
  --remove-bg-noise         Remove background noise from samples

voices - Voice Library

elevenlabs voices list
elevenlabs voices info --id <voice_id>
elevenlabs voices delete --id <voice_id>

sfx - Sound Effects

elevenlabs sfx [options] <description>

Options:
  -d, --duration <seconds>  Approximate duration
  -o, --out <file>          Output file path
  --influence <0-1>         Prompt influence

isolate - Voice Isolation

elevenlabs isolate [options] <audio_file>

Options:
  -o, --out <file>          Output file path

dub - Audio/Video Dubbing

elevenlabs dub [options] <file>

Options:
  -t, --target <lang>       Target language (required)
  -s, --source <lang>       Source language (auto-detected if not specified)
  -o, --out <file>          Output file path
  --status --id <id>        Check dubbing status
  --download --id <id>      Download dubbed audio

Supported Languages:
  en (English), es (Spanish), fr (French), de (German), it (Italian),
  pt (Portuguese), pl (Polish), hi (Hindi), ar (Arabic), zh (Chinese),
  ja (Japanese), ko (Korean), nl (Dutch), ru (Russian), tr (Turkish),
  vi (Vietnamese), sv (Swedish), da (Danish), fi (Finnish), and more...

🔧 Environment Variables

Variable	Description	Default
ELEVENLABS_API_KEY	Your ElevenLabs API key (required)	-
ELEVENLABS_DEFAULT_VOICE	Default voice name	Rachel
ELEVENLABS_DEFAULT_MODEL	Default TTS model	eleven_turbo_v2_5
ELEVENLABS_OUTPUT_DIR	Default output directory	~/.openclaw/audio

📖 Voice Models

Model	Latency	Languages	Best For
eleven_flash_v2_5	~75ms	32	Real-time, streaming
eleven_turbo_v2_5	~250ms	32	Balanced quality/speed
eleven_multilingual_v2	~500ms	29	Long-form, highest quality

🎭 Built-in Voices

• Rachel - Calm and professional female voice
• Adam - Confident male voice
• Antoni - Energetic male voice
• Bella - Soft female voice
• Domi - Strong female voice
• Elli - Warm female voice
• Josh - Deep male voice
• Sam - Young male voice

💰 Pricing

ElevenLabs API pricing (approximate):

• Flash v2.5: ~$0.06/min
• Turbo v2.5: ~$0.06/min
• Multilingual v2: ~$0.12/min
• Voice cloning: Included in plan
• Sound effects: ~$0.02/generation

Free tier: ~10,000 characters/month

🧪 Testing

Run the test suite:

./test.sh <your_api_key>

🔗 Links

• ElevenLabs Dashboard
• API Documentation
• Voice Library
• Pricing

📝 License

MIT License - See OpenClaw project license

🏆 Hackathon Submission

This skill was created for the Clawdbot x ElevenLabs Developer Challenge.

Features:

• ✅ Full ElevenLabs API coverage (TTS, STT, Clone, SFX, Dub, Isolate)
• ✅ OpenClaw-native implementation with SKILL.md
• ✅ Comprehensive CLI with help and error handling
• ✅ Multi-language support (32+ languages)
• ✅ Voice library management
• ✅ Test suite included

Technically Deep:

• Implements all major ElevenLabs APIs
• Proper error handling and rate limit awareness
• JSON processing with jq
• File handling and directory management
• Form data and multipart uploads

Practically Useful:

• Content creators: audiobooks, podcasts, video narration
• Multi-channel users: voice messages across all OpenClaw channels
• Business users: professional voice content
• Accessibility: voice interaction for text-based tasks

Thoughtfully Implemented:

• Follows OpenClaw skill conventions
• Secure API key handling
• Graceful error messages
• Comprehensive documentation
• Easy installation and setup