You can create a common speech profile suitable for both IVR and Voice Bot applications and designate languages and corresponding Text-to-Speech (TTS) and Automatic Speech Recognition (ASR) voices for each language.

Note that any modifications made won't impact previous IVR flows and these adjustments exclusively affect new flows created from this point forward.

To Add a Speech Profile

1. Click the New Tab icon. Under the Modern Care tab, click Voice Care within Listen.

   ​

2. Navigate to the Speech Profiles section from the left pane of the Voice Settings window.

3. Click Create Profile in the top right corner.

4. On the Speech Profile Settings window, add the following details.

   1. Begin by providing a Profile Name and selecting the Default Language. All the languages added to the system will appear here. To add more languages, kindly raise a support ticket at tickets@sprinklr.com. If no set language is detected in IVR, this default language will be sourced from the speech profile, and the corresponding Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) models will be utilized.

   2. Subsequently, include Speech Configs. Pick a Language from the dropdown menu and integrate the preferred Text-to-Speech (TTS) Model along with the desired Playback Speed. Should you wish to incorporate a specific TTS Model into the options, kindly raise a support ticket at tickets@sprinklr.com.

   3. Then, proceed to choose the Speech-to-Text (STT) Model to generate the transcripts.

      

   4. Repeat this process for various languages. Once completed, Save your settings using the bottom-right corner option.

To Add Speech Profile to IVR

While creating an IVR, select the created Speech Profile from the dropdown. When the Interactive Voice Response (IVR) system is deployed, the textual content configured within the IVR will be directed to the Text-to-Speech (TTS) model that corresponds to the language linked with the speech profile.