Configure AI Voicemail Transcription with ElevenLabs Service

Cloud Voice can transcribe voicemail recordings into readable text using the ElevenLabs Speech to Text service. This page walks you through generating an ElevenLabs API (Application Programming Interface) key and connecting it to Cloud Voice so incoming voicemail audio is converted to text automatically.

Tip

If AI Call Transcription is already set up against the ElevenLabs Speech to Text service, you don’t have to re-enter everything. Turn on Voicemail Transcription, choose Copy Settings from Call Transcription to pull in the existing configuration, then adjust whatever you need.

Requirements

Before you begin, confirm your system meets the following:

Item	Requirement
Firmware	Version 84.23.0.83 or later.
Subscription	An Enterprise Plan or Ultimate Plan subscription, which makes the AI Transcription feature available.

Prerequisites

Item	Prerequisite
Network access	Cloud Voice must be able to reach api.elevenlabs.io to call the ElevenLabs Speech to Text service.
ElevenLabs account	An ElevenLabs account with enough transcription-minute quota, along with its username and password.

Note

To confirm the ElevenLabs domain is reachable, run a ping test from the admin portal at Maintenance > Troubleshooting > IP Ping.

Step 1: Create an API key on ElevenLabs

Cloud Voice authenticates each request to the Speech to Text service with an API key, so create one on the ElevenLabs platform first.

1. Sign in to the ElevenLabs Platform with your ElevenLabs username and password, then open Developers > API Keys.

   

2. In the top-right corner of the key list, select Create Key.

3. Fill in the new key in the sidebar that opens.

   

   a. Enter a recognizable Name for the key.

   b. Under Endpoints, select Access next to Speech to text.

   c. Under Administration, select Access next to Models.

   Caution

   Grant both permissions. The key needs the Speech to text endpoint to send voicemail audio and the Models permission (under Administration) to run the transcription model. If either grant is missing, Cloud Voice cannot complete transcription.

   d. Select Create Key in the bottom-right corner. A window appears with the generated key.

   

   e. Select Copy to Clipboard and store the key somewhere safe for the next step.

Step 2: Configure voicemail transcription in Cloud Voice

With the API key in hand, set up the connection in the Cloud Voice admin portal.

1. Sign in to the admin portal and open AI > AI Toolbox.

2. Turn on Voicemail Transcription.

3. Enter the service details.

   

   a. Set Service Type to Custom Service.

   b. Set Transcription Service Provider to Elevenlabs.

   c. Paste the API key you copied from ElevenLabs into the API key field.

4. Under Language, choose the language to detect and transcribe in voicemail messages.

   Note

   Choose Auto Detection to transcribe each audio segment of a mixed-language message in the language it was spoken.

5. In Extension Scope for This Feature, decide who can use voicemail transcription:

   • All Extensions: every extension has access.
   • Specific Extensions: only the extensions, extension groups, or departments you select have access.

6. Select Save.

Result

When Transcription Connection Status shows Enable, voicemail transcription through the ElevenLabs Speech to Text service is active. Both group and extension voicemail recordings are now detected and converted to readable text in the language you chose.

Note

You can set the transcription language per extension at Extension and Trunk > Extension > AI > Voicemail Transcription > Language. That per-extension language applies to the extension’s personal voicemails and to group voicemails delivered in Broadcast to Members mode. For details, see Configure AI Voicemail Transcription for an Extension.