sound-generator

Sound Generator

Generates AI sound effects using ElevenLabs. Describe any sound and hear it instantly. Supports modifying, enhancing, replaying, and adjusting duration.

Trigger Words

• "make a sound"
• "create a sound"
• "sound effect"
• "generate a sound"

Setup

Requires an ElevenLabs API key.

1. Get an API key at elevenlabs.io
2. Replace YOUR_ELEVENLABS_API_KEY_HERE in main.py

How It Works

1. User describes a sound ("thunder in a canyon")
2. API generates the sound effect
3. Audio plays back immediately
4. User can then:
   • Modify: "Make it longer" / "Make it shorter"
   • Enhance: "Enhance it" (LLM improves the prompt)
   • Replay: "Play it again"
   • New sound: Describe something else
   • Exit: "Stop" / "Done"

Commands

Say	What Happens
"thunder in a canyon"	Generates new sound
"make it longer"	Regenerates at +3 seconds
"make it shorter"	Regenerates at -2 seconds
"enhance it"	LLM improves the prompt, regenerates
"again" / "replay"	Replays last sound
"stop" / "done"	Exits the Ability

Example Conversation

User: "Make a sound" AI: "Sound generator ready. Describe what you want to hear." User: "Rain on a tin roof" AI: "Creating rain on a tin roof." [Sound plays] AI: "There we go. You can modify it or ask for a new sound." User: "Make it longer" AI: "Regenerating at 8.0 seconds." [Longer version plays] User: "Done" AI: "Closing sound generator."