🔊 AI Sound Alerts
The AI Sound Alerts widget lets viewers trigger sound effects on your stream, including preset sounds and AI-generated audio effects powered by ElevenLabs.
Overview
Viewers use the !sound command to play sound effects on stream. Choose from preset sounds or enable AI-generated sounds for unlimited creativity.
Default Cost: 50 Juice (global)
How It Works
- Viewer types:
!sound airhornor!sound generate explosion - Juice is deducted from their balance
- Sound plays on your Sound Alerts overlay
- Optional visual animation appears
Sound Types
Preset Sounds
Pre-uploaded sound effects that you configure in advance:
- Upload your own sound files
- Set individual costs per sound
- Enable/disable specific sounds
AI Generative Sounds (ElevenLabs)
Create any sound effect on demand using AI:
- Type a description, get a unique sound
- Powered by ElevenLabs technology
- Must be enabled separately in settings
Settings
Basic Settings
| Setting | Description | Default |
|---|---|---|
| Global Cost | Default Juice cost for all sounds | 50 |
| Enabled | Master toggle for Sound Alerts | On |
| Generative Enabled | Enable AI-generated sounds | Off |
| Broadcast to Chat | Announce sounds in chat | Off |
Individual Sound Settings
Each preset sound can have:
- Custom cost (overrides global)
- Enabled/disabled status
- Custom display name
Setup Instructions
Step 1: Enable Sound Alerts
- Go to Widgets → Sound Alerts in your dashboard
- Ensure the Enabled toggle is ON
- Set your global Juice cost
Step 2: Add Preset Sounds
- Click Add Sound in the Sound Alerts settings
- Upload an MP3 or WAV file
- Set a name (this is the command trigger)
- Optionally set a custom cost
- Click Save
Step 3: Add Overlay to OBS
- Copy the Sound Alerts overlay URL
- In OBS, add a Browser Source
- Paste the URL
- Set dimensions: 800 x 600 (recommended)
- Enable audio output from the browser source
Step 4 (Optional): Enable AI Sounds
- In Sound Alerts settings, toggle Generative Enabled to ON
- AI sounds will use the global cost
Commands
| Command | Description |
|---|---|
!sound <name> | Play a preset sound |
!sound generate <description> | Generate an AI sound |
!sounds | List available sounds |
Examples
!sound airhorn
!sound bruh
!sound generate dramatic thunder
!sound generate crowd cheeringManaging Sounds
Uploading Custom Sounds
- Go to Widgets → Sound Alerts
- Click Add Sound or the upload button
- Select an audio file (MP3, WAV, OGG supported)
- Enter a trigger name (no spaces, alphanumeric)
- Set the cost and enable/disable status
- Save changes
Organizing Sounds
- Use clear, memorable names for triggers
- Disable sounds that aren't popular
- Set higher costs for longer or premium sounds
Removing Sounds
- Find the sound in your list
- Click the delete/remove button
- Confirm deletion
AI Generated Sounds
When Generative Enabled is ON, viewers can create unique sounds:
How It Works
- Viewer types:
!sound generate car crash - ElevenLabs AI generates the audio in real-time
- Sound plays on your overlay
- Each generation is unique!
Best Practices for AI Sounds
- AI sounds take 2-5 seconds to generate
- Results vary based on the description
- Some descriptions work better than others
- Consider setting a higher cost for AI sounds
Content Considerations
The AI has built-in content filters, but consider:
- Monitoring AI sound descriptions
- Using word filters to block inappropriate prompts
- Setting higher costs to reduce misuse
Troubleshooting
Sounds Not Playing
| Issue | Solution |
|---|---|
| No audio | Check browser source has audio enabled |
| "Sound not found" | Verify the sound name is correct |
| Command ignored | Check widget is enabled |
| Delayed playback | Normal for AI-generated sounds (2-5 sec) |
Upload Issues
- Ensure file is under 5MB
- Use supported formats (MP3, WAV, OGG)
- Avoid special characters in filenames
AI Sounds Not Working
- Verify Generative Enabled is ON
- Check the description isn't triggering content filters
- Try a different description
Best Practices
- Curate Your Sounds: Quality over quantity
- Balance Costs: Consider stream pacing
- Preview Before Enabling: Listen to all sounds first
- Use Descriptive Names: Make it easy for viewers to remember
- Set Cooldowns: Prevent spam (use bot moderation)
FAQ
Q: What audio formats are supported? A: MP3, WAV, and OGG files up to 5MB.
Q: Can viewers see what sounds are available? A: Yes, the !sounds command lists all enabled sounds.
Q: How long can sounds be? A: Recommended under 30 seconds for best viewer experience.
Q: What's the difference between global and individual costs? A: Global cost is the default; individual costs override it for specific sounds.
Q: Are AI-generated sounds saved? A: No, each AI sound is generated fresh and not stored.
