Generating Audio
This guide covers how to assign AI voices to your characters and generate audio productions from your screenplay.
Assigning Voices
Browse the Voice Library
- Go to the Characters tab
- Click on a character name
- The voice library panel will open
The voice library includes:
- Apple Text-to-Speech (Free tier): Quality voices built into your device
- ElevenLabs voices (Pro tier): Premium AI voices via your ElevenLabs API key (standard ElevenLabs API generation charges apply)
Preview Voices
Before assigning a voice:
- Click the Play button next to any voice
- Listen to a sample of how that voice sounds
- Try different voices until you find the right fit
Assign a Voice
- Select the voice you want
- Click Assign to apply it to the character
- The voice will be used for all that character’s dialogue
Voice Settings
Fine-tune each character’s voice:
- Speed: Adjust speaking pace (0.5x to 2x)
- Pitch: Modify voice pitch
- Emotion: Select emotional tone (where supported)
Generating Audio
Generate a Single Scene
- Go to the Scenes tab
- Click on the scene you want to generate
- Click the Generate button
- Wait for processing to complete
- Listen to the result
Generate Multiple Scenes
- Select multiple scenes using checkboxes
- Click Generate Selected
- Scenes will be processed in order
- Progress is shown in the status bar
Batch Generation
For full screenplay generation:
- Click Generate All
- All scenes will be queued for processing
- Generation happens in the background
- You can continue working while it processes
Reviewing Results
After generation:
- Play the scene to listen
- Compare with the original dialogue
- Regenerate if needed with different settings
Making Adjustments
If a scene doesn’t sound right:
- Try adjusting voice speed or pitch
- Switch to a different voice
- Regenerate the scene
Tips for Best Results
Voice Selection
- Match voice characteristics to character descriptions
- Consider age, gender, and personality
- Preview multiple options before deciding
Pacing
- Adjust speed for emotional scenes
- Action scenes may benefit from faster pacing
- Dramatic moments often work better slower
Generation Quality
- Shorter scenes generate more consistently
- Very long dialogue may need to be split
- Complex character names may affect pronunciation
Voice Providers
Free tier: Apple Text-to-Speech voices (built into your device, works offline)
Pro tier: ElevenLabs premium voices via BYOK (Bring Your Own Key). Standard ElevenLabs API generation charges apply.