Skip to main content
Resonance’s text-to-speech engine transforms written text into high-quality, natural-sounding audio using advanced AI voice models.

​
Before You Start

You need an active subscription to generate speech. See the Usage & Billing guide for pricing details.

​
Generating Your First Audio

1

Navigate to Text to Speech

From the sidebar, click Text to Speech or visit /text-to-speech in your dashboard.
2

Enter Your Text

In the main text input panel, type or paste the text you want to convert to speech.
Maximum text length: 5,000 characters per generation
3

Select a Voice

Click the Voice Style dropdown to choose from:
  • Team Voices - Custom voices created by your organization
  • Built-in Voices - Pre-configured system voices
Each voice displays its name and category (e.g., Audiobook, Conversational, Narrative).
4

Adjust Voice Settings (Optional)

Fine-tune how the voice sounds using the sliders:
Controls variation in voice expression.
  • Consistent (0.0): Predictable, stable delivery
  • Expressive (2.0): More dynamic and varied
  • Default: 0.8
Affects the range of possible voice outputs.
  • Stable (0.0): Minimal variation
  • Dynamic (1.0): Maximum variety
  • Default: 0.95
Controls the breadth of expressive choices.
  • Subtle (1): Limited expression
  • Dramatic (10,000): Full expressive range
  • Default: 1,000
Prevents repetitive patterns in speech.
  • Rhythmic (1.0): Allows repetition
  • Varied (2.0): Penalizes repetition
  • Default: 1.2
5

Generate Audio

Click the Generate button to create your audio.The system will:
  1. Validate your active subscription
  2. Process your text with the selected voice
  3. Apply your voice settings
  4. Generate and store the audio file
  5. Record usage for billing (charged per character)
6

Preview and Download

Once generated, you’ll be redirected to the audio detail page where you can:
  • Play the audio with the built-in player
  • Download the audio file
  • View generation settings
  • See character count and metadata

​
Quick Text Suggestions

Resonance provides prompt suggestions to help you get started quickly. Click any suggestion to populate the text input with sample content.

​
Generation History

All your generated audio files are saved and accessible through:
  • The History drawer (icon in the sidebar)
  • The main dashboard
Your generation history is scoped to your organization, so all team members can access previously generated audio.

​
Usage Metering

Each generation is metered based on character count:
  • Characters are counted from your input text
  • Usage is tracked in real-time
  • Costs accumulate throughout your billing period
  • View estimated costs in the sidebar under “Current usage”

Pricing

Learn about character-based pricing

Custom Voices

Create your own voice clones

​
Error Handling

​
Subscription Required

If you don’t have an active subscription, you’ll see: 
Subscription required
[Subscribe]
Click Subscribe to start a checkout session with Polar.

​
Voice Not Found

If the selected voice is unavailable (e.g., a team voice was deleted), select a different voice from the dropdown.

​
Generation Failed

If generation fails:
  1. Check your internet connection
  2. Verify your text is under 5,000 characters
  3. Try a different voice
  4. Contact support if the issue persists

​
Best Practices

  • Use voices in the Audiobook or Narrative category
  • Keep temperature lower (0.6-0.8) for consistency
  • Use higher repetition penalty (1.3-1.5) for natural flow
  • Select Podcast or Conversational voices
  • Increase temperature (0.8-1.0) for more expression
  • Use moderate Top P (0.9-0.95) for natural variation
  • Choose Customer Service category voices
  • Lower temperature (0.5-0.7) for professional tone
  • Keep settings stable for brand consistency
  • Use Characters category
  • Higher temperature (1.0-1.5) for dramatic delivery
  • Adjust expression range based on character needs

​
Next Steps

Build docs developers (and LLMs) love

Get started for freeTalk to us