Salesforce Docs Portal

Configure Speech-to-Text Model

The speech-to-text model determines how spoken audio is processed and transcribed during voice interactions.

Required Editions

This article applies to:

Salesforce Voice (Native Telephony)

Available in: Agentforce Contact Center with Salesforce Voice
Available in: Enterprise, Unlimited, and Developer Editions

To configure the speech-to-text model at the organization level.

  1. From Setup, use Quick Find to go to Speech-to-text model setting.
  2. Select the model from the list.
    • Low Latency Speech Model
    • Universal Speech Model