Configure Voice Model on Geta.ai

Steps to Configure Voice

1. Navigate to Voice Settings

  • Log in to your Geta.ai dashboard.

  • From the sidebar, go to the Chatbot Builder module.

  • Select your Voice Bot from the list.

  • Click the Voice tab under Agent Setup.


2. Select a Voice

  • Click Select Voice to open the voice selection panel.

  • Use the Search option to find a specific voice.

  • Click Select Language to choose from multiple language options.

  • Choose your preferred language.

  • Click Select Gender to filter voices by male or female.

  • Click the Audio Sample icon to listen to or download the sample.

  • Close the window once you’ve found the voice you want to use.


3. Enable Flash Model (Recommended)

  • Turn on the Use Flash toggle.

    • Flash is the recommended low-latency voice model for smoother call experiences.


4. Select TTS Output Format

  • Click on the TTS Output Format dropdown.

  • Choose the audio format your organization prefers (e.g., ÎĽ-law 8000 Hz).


5. Add Pronunciation Dictionaries

  • Click Add Pronunciation Dictionaries.

  • Enter a Dictionary Name.

  • Add the Alias (word or acronym) and the String to Replace (how the bot should pronounce it).

    • Example: Alias: “SCIT” → Replace with: “Ess-See-Eye-Tee”.


6. Adjust Voice Controls

Fine-tune your agent’s audio behaviour:

  • Optimize Streaming Latency – Reduces delay in voice responses.

  • Stability – Controls voice consistency across calls.

  • Speed – Adjusts how fast the voice speaks.

  • Similarity – Controls how closely the output matches the selected voice tone.


7. Save Your Settings

  • Click Save Changes to apply all voice configurations.



Next Steps

  • Test your Voice Bot by making a sample call.

  • Move to the Analysis module to view call metrics.

  • Visit Advance Settings to configure more detailed voice behavior.


Was this article helpful?
© 2025 Geta.ai