Skip to main content

​
Introduction

The AI Voiceover feature in Klyra AI unlocks advanced text-to-speech capabilities, helping you generate realistic narrations, ads, training audio, and more. With access to six AI vendors — Elevenlabs, AWS, GCP, Azure, OpenAI, and Speechify — you can synthesize voices in thousands of variations across 140+ languages and dialects. This makes it a versatile tool for educators, marketers, podcasters, accessibility providers, and content creators.

​
How to Use

🔎 Navigation: Dashboard → Left Panel → AI Audio & Voice → AI Voiceover
Klyra Ai Voiceover Dashboard Web
1

Pick your language

  • Select from 140+ supported languages and dialects.
  • Example: English (US), Spanish (MX), French (FR), Hindi (IN).
2

Choose your voice

  • Browse available voices from the supported AI vendors.
  • A 🔊 voice preview button is available to audition vendor samples.
3

Add your script details

  • Title: Enter a title to help organize your script.
  • Script: Paste or write the text you want converted into speech.
  • Some vendors (AWS, GCP) support SSML (Speech Synthesis Markup Language) tags for advanced control (pauses, emphasis, speed, etc.).
4

Select output format

  • Choose the audio format best suited to your use case:
    • MP3 — lightweight, web-friendly.
    • WAV — uncompressed, studio quality.
    • OGG / WEBM — optimized for web applications.
5

Listen or Synthesize (consumes credits)

  • After entering your script and format, you’ll see two options:
    • Listen — Quickly renders your current script so you can audition the result.
    • Synthesize — Produces the final voiceover file in your chosen format.
Both Listen and Synthesize consume Character credits based on the script length.
👉 To save credits, try Listen with a short sample before synthesizing the full script.
6

Access your results

  • Your generated voiceover appears in AI Voiceover Results below the workspace.
  • It is also stored under AI Panel → Documents → All Voiceovers for future reference.
7

Enhance with SSML (Optional)

  • For vendors like AWS and GCP, you can insert SSML tags into your script.
  • Examples:
    • <break time="1s"/> → Adds a pause.
    • <emphasis level="strong">important</emphasis> → Stresses a word.
    • <prosody rate="slow">This part is slower</prosody> → Adjusts pacing.
8

🎧 Sample Output

Here’s a sample voiceover generated using Klyra AI Voiceover:
This demonstrates how natural and professional Klyra AI’s text-to-speech sounds when converting written content into lifelike audio.

​
💡 Tips

  • Start with shorter scripts to test different voices and tones without using too many credits.
  • Match voice to audience: A calm, clear voice works well for e-learning; a lively, energetic one is great for ads.
  • Use SSML sparingly: It enhances realism but should feel natural.
  • Choose format wisely: Use WAV for professional editing, MP3 for distribution.
  • Organize by title: Helps you quickly locate saved scripts in your documents.

​
📌 Example Use Cases

  • E-learning: Convert training material into audio lectures for global learners.
  • Marketing & Ads: Produce multilingual ad campaigns with professional narrations.
  • Content Creators: Generate narration for YouTube videos, podcasts, or audiobooks.
  • Accessibility: Enable voice-read articles, blogs, and documents.
  • Corporate Training: Deliver compliance and onboarding audio in multiple languages.