Text-to-Speech

Text-to-Speech (TTS) lets you create professional audio prompts by typing text instead of recording audio. AI generates natural-sounding speech in your chosen voice, ready for IVR menus, queue announcements, voicemail greetings, and more.

Generating Prompts

1. Navigate to IVR Builder or Admin > Voice Files
2. Click Generate from Text
3. Type your prompt text
4. Select a voice — choose from multiple voices and styles
5. Select the language
6. Click Generate
7. Preview the result — listen before saving
8. Save to your prompt library

Note

Generated audio is stored as standard WAV files in the PBX sounds directory, making them available to all IVR menus and call flows immediately.

Where TTS Is Used

Use Case	Example
IVR Menus	”Press 1 for Sales, Press 2 for Support”
Queue Announcements	”Your estimated wait time is 3 minutes”
Voicemail Greetings	”You’ve reached our sales department…”
Voicemail Drop	Pre-recorded messages left on answering machines via AMD
Hold Messages	”Thank you for holding, an agent will be with you shortly”
After-Hours Messages	”Our office is currently closed. Please call back…”

Configuration

Navigate to Admin > AI Settings > TTS:

Setting	Description
Provider	TTS service provider
Voice	Default voice selection from the provider’s voice library
Model	TTS model variant (standard, enhanced, multilingual)

Voice Selection

The TTS provider offers a wide library of voices with different characteristics:

• Gender — male, female, and neutral options
• Accent — American, British, Australian, and more
• Tone — professional, friendly, authoritative, warm
• Language — multilingual support for international deployments

Best Practices

Tip

• Preview before saving — always listen to the generated audio to ensure it sounds natural and clear
• Keep prompts concise — shorter messages are easier for callers to follow and reduce wait times
• Use professional tone — match the voice and style to your brand identity
• Test with callers — verify prompts sound clear over phone audio quality (not just through computer speakers)
• Version your prompts — generate new versions rather than overwriting, so you can revert if needed

Caution

TTS usage is billed per character by the provider. Long prompts and frequent regeneration will increase costs. Preview carefully before generating final versions.