Auto-Talk-TTS Skill#

Overview#

Automatically speaks every message you generate using Microsoft Edge's neural TTS service. Runs asynchronously in the background so your conversation continues immediately while audio generates.

Quick Start#

Every message you send is automatically spoken aloud.

The skill wraps your output with auto-speak which:

1. Installs node-edge-tts if needed
2. Converts your message to speech asynchronously
3. Plays the audio in the background
4. Continues your conversation immediately

Usage#

Automatic: Every message is auto-spoken. No action needed from you.

Manual override: If you need to speak something specific:

bash

auto-speak "Your text here"

How It Works#

1. Detect output: When you generate a message
2. Wrap with auto-speak: The message gets passed through the auto-speak wrapper
3. Install if needed: First run installs node-edge-tts package
4. Generate audio: Convert text to MP3 in background
5. Play audio: Use afplay to play the audio file
6. Continue: Your conversation flows without waiting for audio

Configuration#

Edit config.json to customize:

json

{
  "voice": "en-US-MichelleNeural",
  "rate": "default",
  "pitch": "default",
  "volume": "default",
  "summaryLength": 100
}

Options:

• voice: Any Edge neural voice (see edge-tts skill for full list)
• rate: Speech speed ( -20% to +50%, or default)
• pitch: Voice pitch ( -20Hz to +20Hz, or default)
• volume: Audio volume ( -100% to +100%, or default)
• summaryLength: How many characters to summarize (default: 100)

Installation#

First run will automatically install node-edge-tts:

bash

npm install -g node-edge-tts

Or use the bundled installer:

bash

cd /Users/stefano/.openclaw/workspace/skills/auto-talk-tts
npm install

Files#

auto-speak#

Wrapper script that:

• Checks if node-edge-tts is installed
• Installs it if needed
• Converts text to speech asynchronously
• Plays audio with afplay
• Continues immediately

config.json#

User configuration for voice, rate, pitch, volume settings.

package.json#

NPM package with node-edge-tts dependency.

SKILL.md#

This file.

Integration with SOUL.md#

Add this line to the top of your SOUL.md:

markdown

## MUST SPEAK WITH TTS EVERY TIME (ASYNCHRONOUSLY)

**I must use the edge-tts skill to convert my response to speech before sending it to you.** Every single message I send must be spoken aloud using the tts tool. This is non-negotiable - I speak to you, I don't just type.

**Async approach:** I run the TTS command in the background using `&` or `nohup` so the conversation continues immediately while audio generates in the background.

Workflow#

Automatic (Recommended)#

1. You generate a message
2. auto-speak wraps it: The message gets passed through the wrapper
3. Background TTS: Audio generates in background process
4. Immediate reply: Your conversation continues without waiting
5. Audio plays: User hears your response

Manual (If needed)#

bash

# Auto-speak any text
auto-speak "Your message here"

# Custom voice
auto-speak "Your message" --voice en-GB-RyanNeural

# Custom speed
auto-speak "Your message" --rate +20%

Testing#

bash

# Test the auto-speak wrapper
auto-speak "This is a test. If you can hear this, it's working!"

# Check if node-edge-tts is installed
npm list -g node-edge-tts

# Install manually if needed
npm install -g node-edge-tts

Troubleshooting#

• No audio? Check that node-edge-tts is installed: npm list -g node-edge-tts
• Install needed? First run will auto-install, or run: npm install -g node-edge-tts
• Audio not playing? Check system volume and audio output settings
• Slow? Use async mode - audio generates in background, conversation continues immediately

Notes#

• Non-blocking: Conversation flows while audio generates
• No API key: Uses Microsoft Edge's free TTS service
• Neural voices: High-quality, natural-sounding speech
• Multiple voices: See edge-tts skill for full voice list
• Configurable: Customize voice, speed, pitch, volume in config.json
• Automatic installation: First run installs node-edge-tts if missing

See Also#

• edge-tts - Core TTS engine
• speak-summary - Alternative auto-speak implementation