Post-Call Text Insights Now in GA. Featuring AI-Generated Summaries

We’re excited to announce a new enhancement to the Post-call Transcription feature of the Vonage Video API – introducing Post-Call Text Insights, now available in GA!

As part of this release, developers can now access AI-generated summaries of recorded video sessions, giving applications smarter, faster ways to surface meaning from conversation transcripts.

What Is Post-Call Text Insights?

Post-Call Text Insights is a new feature in the Vonage Video API transcription workflow that adds contextual data to archived session transcripts available by enabling transcription and summary when starting an archive. It currently includes:

• Summary: An AI-generated high-level description of the conversation, giving a concise overview of what was discussed during the session.

Instead of parsing full transcripts manually, your application can now present users with an instant overview of their call – ideal for telehealth, education, recruiting, and more.

Example Use Case

Let’s say your app records a 20-minute video consultation. With Post-Call Text Insights, you can automatically extract a short summary like:

“The doctor and patient discussed symptoms of seasonal allergies. A new prescription was recommended, and a follow-up was scheduled.”

This makes it faster for users (and systems) to understand what happened – no need to scan the full transcript.

How to Enable Summary

To include a Summary in the transcription file, simply pass the new transcriptionProperties object when starting the archive with a hasSummary property (Boolean) to include an AI-generated summary in the transcription. The default value for hasSummary is false (the transcription summary is not included).

When you use the Vonage Video REST API start archiveAPI call, set the hasAudio and hasTranscription properties to true in the JSON properties you sent to the start archive REST method:

The resulting transcription file will now include a summary field along with the full, speaker-tagged transcript.

Refer to the Transcription Guide for full documentation.

An example of enabling transcription from a CURL request:

application_id="12345abc"

json_web_token="jwt_string" # replace with a JSON web token

data='{
  "sessionId": "1_MX40NzY0MDA1MX5-fn4",
  "hasAudio": true,
  "hasVideo": true,
  "hasTranscription": true,
  "transcriptionProperties": {
    hasSummary: true
  },
  "name": "archive_test",
  "outputMode": "individual"
}'

curl \
  -i \
  -H "Content-Type:application/json" \
  -X POST \
  -H "Authorization: Bearer $json_web_token" \
  -d "$data" \
  https://video.api.vonage.com/v2/project/$application_id/archive

Why This Matters

With Post-Call Text Insights, you can:

• Highlight important points from calls to lessen the mental effort required to review them.

• Enable follow-up processes or systems to act on call content.

• Save time reviewing archived calls

• Build smarter UI/UX around session history.

Try It Out

The Summary capability within Post-Call Text Insights is now available in GA for all Vonage Video API accounts. Just pass the new parameter to begin using the feature – no approval needed!

We’d Love Your Feedback!

This is just the beginning of making Post-Call Transcriptions even better. We’re actively working on expanding Post-Call Text Insights with more features, and your feedback will help shape what comes next.

Try it out and let us know how it works for your use case. Join the Vonage Community Slack and tell us your experience with the feature.

Let us know what you'd like to see next!