Configure AI Voicemail Transcription with Google Service

Yeastar P-Series Software Edition supports AI voicemail transcription powered by Google Cloud Speech-to-Text service (API version: V2), converting audio in voicemail messages into readable text. This topic describes how to configure AI voicemail transcription with the third-party service on the PBX.

Tip: If you have already configured AI Call Transcription with Google Cloud Speech-to-Text service, you can enable Voicemail Transcription and click Copy Settings from Call Transcription to reuse the configuration, then modify the settings as needed.

Requirements

Yeastar P-Series Software Edition should meet the following requirements:
Item Requirement
Firmware 83.23.0.83 or later.
Subscription Subscribe to Enterprise Plan or Ultimate Plan to ensure the AI Transcription feature is available.

Prerequisites

Item Prerequisites
PBX network access Make sure the Yeastar P-Series Software Edition can access the following domains to use Google Cloud Speech-to-Text service:
  • oauth2.googleapis.com
  • <region>-speech.googleapis.com
  • <region>-storage.googleapis.com
Note:
  • Select the desired region from the list below and replace <region> with it (e.g., us-speech.googleapis.com). It is recommended to select a region closest to your PBX deployment location to reduce network latency and ensure stable transcription.
    • us: United States
    • eu: Europe
    • asia-southeast1: Singapore
    • asia-northeast1: Tokyo

  • You can verify domain accessibility on the PBX (Path: Maintenance > Troubleshooting > IP Ping). If the PBX can not access the domains, go to System > Network > Basic Settings to configure the PBX network.
Third-party service account Prepare a Google account with sufficient transcription minute quotas for Google Cloud Speech-to-Text service and permissions to manage Google Cloud Storage (GCS) buckets, and obtain the username and password.

Procedure

Step 1. Create an API key for Google Cloud Speech-to-Text service

To securely access the Google Cloud Speech-to-Text service from the PBX, you must create an API key on Google Console first and export it as a JSON file used to authenticate the service API request.

  1. Log in to Google Console using your Google username and password.
  2. Create a new project.
    1. At the top-left corner, click current project tab, and then click New project in the pop-up window.

    2. In the New Project page, create a new project.

      1. In the Project name field, enter a name to identify the project.
      2. Optional: Click Browse to select the desired organization.
      3. Click Create.
  3. In the new created project, enable the Cloud Speech-to-Text API service.

    1. Go to APIs & Services > Library, search for "Cloud Speech-to-Text API" in the library.
    2. In the search result list, click the Cloud Speech-to-Text API card to enter its product details.
    3. Click Enable.

      This service displays with Enabled status.

  4. Create a service account for the new created project.
    1. Go to IAM & Admin > Service accounts, click Create service account at the top navigation bar.

    2. Create a service account.

      1. In the Service account name field, enter a name to identify the service account.
      2. Click Create and continue.
      3. In the Role drop-down list, select Owner.
      4. Click Continue.
      5. Click Done.

        The Service accounts list displays the created service account.

  5. Create an API key and generate its JSON file for the new created service account.

    1. On the Service accounts page, click beside the created service account, and select Manage keys.
    2. On the Keys page, click Add key and select Create new key.

      A key type selection window pops up.

    3. In the Key type section, select JSON type, and click Create.

      The system automatically downloads the JSON file that contains the API key to your computer. You can check the JSON file on your computer and save it for later use.

Step 2. Configure AI voicemail transcription on the PBX

After you create an API key for Google Cloud Speech-to-Text service, you need to configure corresponding settings on the PBX to establish a connection with the service API.

  1. Log in to PBX web portal, go to AI > AI Toolbox.
  2. Turn on the switch of Voicemail Transcription.
  3. Configure AI voicemail transcription service.

    1. In the Service Type drop-down list, select Custom Service.
    2. In the Transcription Service Provider drop-down list, select Google.
    3. Click Browse to upload the API key JSON file.
    4. In the Region drop-down list, select the desired region.
      • us (multi-region): United States
      • eu (multi-region): Europe
      • asia-southeast1: Singapore
      • asia-northeast1: Tokyo
  4. In the Language drop-down list, select the desired language to be detected and transcribed for voicemail messages.
  5. In the Extension Scope for This Feature section, specify which extensions / extension groups / departments are allowed to use voicemail transcription feature.
    • All Extensions: All extensions can use this feature.
    • Specific Extensions: Only selected extensions can use this feature.
  6. Click Save.

Result

The Transcription Connection Status displays Enable, indicating that the AI voicemail transcription feature powered by Google Cloud Speech-to-Text service is enabled. Both group and extension voicemail messages can be detected and transcribed into readable text in the specified language via the Google service.

Note: You can customize transcription language for single extension (Path: Extension and Trunk > Extension > AI > Voicemail Transcription > Language), which will be applied to the group voicemails in Broadcast to Members mode and personal voicemails delivered to the extension. For more information, see Configure AI Voicemail Transcription for an Extension.