Skip to main content
Platform voices are Retell’s curated voice library, fine-tuned specifically for conversational AI over the phone. They handle fillers, pacing, and conversational rhythm well, and are calibrated for clarity at telecom bitrates. When using a platform voice, fallback is handled automatically to maintain the same voice experience — you do not need to configure a TTS fallback plan in Security & fallback settings.
If you are using a non-platform voice, you will need to configure TTS fallback manually.

Select a platform voice

Use the voice selector in your agent settings to browse platform voices, filter by gender or accent, preview with the play button, and save your choice. Each voice has a voice ID in the form retell-<Name> (for example, retell-Cimo) that you can pass as voice_id when you create or update an agent via the API.
1

Open the voice selector

Open an agent. In Agent settings, click the voice chip in the Voice & Language row.
Agent settings panel showing the Voice and Language row with the Cimo voice chip
2

Browse Platform Voices

The Platform Voices tab opens first. Use the Gender and Accent filters, or search by name, to narrow the list. Press play on a card to preview, then select a voice and click Save.
Select Voice dialog on the Platform Voices tab, with Add voice clone, Gender and Accent filters, Search, and the platform voice grid
Start with one of these platform voices based on accent, age, and gender. Preview a few with the play button before you choose — the attributes below match what the dashboard shows on each card. Grouped by need:
  • American female, middle aged: Cimo, Brynne, Kate
  • American female, young: Chloe, Della, Rita
  • American male: Nico (middle aged), Leland (young)
  • British: Willa (middle aged), Maren (young)
  • Mexican accent: Andrea, Claudia (middle aged); Gaby (young); Alejandro (young male)

Custom voice cloning

You can clone a voice as a platform voice from the dashboard or the API. Platform clones get the same automatic TTS fallback as built-in platform voices.

From the dashboard

On the Platform Voices tab, click Add voice clone. Enter a voice name, upload an audio or video clip (up to 10 MB), confirm the consent checkbox, then click Save.
Add Voice Clone dialog with Voice Name, upload audio clip up to 10 MB, and consent checkbox

From the API

You can also clone your own voice as a platform voice by setting voice_provider to platform when calling the Clone Voice API. The cloned voice will behave the same as a built-in platform voice — fallback is handled automatically and you do not need to configure TTS fallback separately.