Configure AI Voicemail Transcription with OpenAI

Cloud Voice can convert the audio in voicemail messages into readable text using OpenAI’s Speech to Text service. This page shows you how to link the phone system to that third-party service and choose which extensions the feature covers.

Tip

Already set up AI Call Transcription against OpenAI Speech to Text? Turn on Voicemail Transcription and click Copy Settings from Call Transcription to bring that configuration over, then change only what differs.

Requirements

Your phone system must meet the following before you begin.

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

Before you begin

Network access

The phone system must be able to reach api.openai.com to use OpenAI Speech to Text.

Note

You can confirm the domain is reachable from the phone system at Maintenance > Troubleshooting > IP Ping.

Third-party account

Have an OpenAI account with enough transcription-minute quota ready, along with its username and password.

Caution

This feature runs on your own OpenAI account and draws down its transcription-minute quota each time a voicemail is transcribed. Confirm the account has enough quota before you turn it on: without it, transcription cannot run. Watch usage closely if you enable the feature for every extension, since every voicemail then consumes quota.

Step 1. Create an API key for OpenAI Speech to Text

So the phone system can authenticate its requests to OpenAI Speech to Text, generate an API key (Application Programming Interface key, the credential that identifies your account) on the OpenAI Platform first.

1. Sign in to OpenAI Platform with your OpenAI username and password, then open API Keys.

2. Above the list of keys, on the right, choose Create new secret key.

3. Fill in the dialog that opens.

   

   a. In Name, enter something that identifies the key.

   b. From the Project drop-down, select the project you want.

   c. Leave the permission set to its default of All, then choose Create secret key. The generated key appears in a new dialog.

   

   d. Choose Copy and store the key somewhere safe for the next step.

Step 2. Configure voicemail transcription on the phone system

With the API key in hand, enter the details on the phone system to establish the connection to OpenAI Speech to Text.

1. Sign in to the phone system web portal and open AI > AI Toolbox.

2. Turn on the Voicemail Transcription switch.

3. Set up the transcription service.

   

   a. From the Service Type drop-down, select Custom Service.

   b. From the Transcription Service Provider drop-down, select OpenAI.

   c. In the API key field, paste the key you created on the OpenAI Platform.

   d. From the Transcription Model drop-down, pick the OpenAI Speech to Text model you prefer. For how the models differ, see OpenAI’s list of Realtime & audio models.

4. From the Language drop-down, choose the language to detect and transcribe in voicemail messages.

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

   • All Extensions: every extension can use the feature.
   • Specific Extensions: only the extensions, extension groups, or departments you select can use it.

6. Click Save.

Result

Transcription Connection Status shows Enable, confirming that AI voicemail transcription via OpenAI Speech to Text is live. Both group and extension voicemail messages are now detected and transcribed into text in your chosen language.

Note

You can set the transcription language for an individual extension from 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. See Configure AI Voicemail Transcription for an Extension for details.