Stay organized with collectionsSave and categorize content based on your preferences.
This page documents production updates to Insights. You can
periodically check this page for announcements about new or updated features,
bug fixes, known issues, and deprecated features.
You can see the latest product updates for all of Google Cloud on theGoogle Cloudpage, browse and filter all release notes in theGoogle Cloud console,
or programmatically access release notes inBigQuery.
To get the latest product updates delivered to you, add the URL of this page to yourfeed
reader, or add thefeed URLdirectly.
August 15, 2025
Quality AI offers the virtual agent platform to help with quality management for virtual agents. Thevirtual agent platformpresents data for the following metrics, aggregated across all an agent's conversations:
Rate of escalation to a human agent
Latencies with respect to operations that an agent performs
Customer sentiments
Conversational Insights offers conversationdatasetsin preview. Choose from your existing history or upload new conversations to curate customizeddatasetsto test and evaluate the results of Agent Assist summarization.
July 18, 2025
Quality AI offersmultiple scorecardsin preview. Withmultiple scorecards, Quality AI can evaluate a single conversation against different criteria and provide multiple conversation scores. You can also choose which scorecard data to view on each page of the console.
June 27, 2025
Quality AI offers theagent engagement platformwhere contact center personnel can visualize agent performance data, including an AI-generated summary. The agent engagement platform also identifies areas where an agent excels and needs help compared to their peers.
Qualilty AI offers agentassessmentsbased on individual conversations. Human evaluators can also add notes to these assessments with feedback for the agent.
Quality AI offerssampling rulesthat filter conversations to reduce the workload for human evaluators, saving contact centers time and money.
Add your owncustom tagsinQuality AI. In addition to the default Business, Compliance and Customer tags, you can apply custom tags to any question in any scorecard.
Conversational Insights offers adevkitto help developers and maintainers perform a wide range of actions, including the following:
Import one or more conversations with metadata.
Transcribe mono audio files.
Create recognizers with STT V2.
Transform transcript data formats from Genesys Cloud or AWS.
Change Conversational Insights global settings.
June 20, 2025
Conversational Insights offerssentiment analysisin GA.Sentiment analysisdetermines the mood within a conversation and assigns a score: positive, neutral, or negative.
May 29, 2025
Conversational Insights offers amerged analysis, which displays the previous results of each analysis type alongside your most recent analysis result. Merged analysis eliminates the need to run every analysis multiple times.
You can integrate Agent Assistsummarization generatorswith Conversational Insights.Summarizationuses existing LLM generators to automatically summarize conversations. You can then export those summaries along with your other Insights data.
March 06, 2025
Conversational Insights offersRule-based analysisas a GA feature to customize your conversation analyses.Rule-based analysisprovides the following customizations:
Filter conversations.
Select a percentage of your dataset.
Designate different types of analysis.
January 28, 2025
Customer Managed Encryption Key (CMEK)is available in GA for all Agent Assist features.CMEKlets you control details of your data encryption keys, view audit logs, and control key lifecycles.
December 23, 2024
Conversational Insights offers LLM-poweredtopic inferenceas a GA feature.Topic inferenceallows you to use your topic model to analyze new conversations and identify topics in real time.
This feature is only available for English.
December 20, 2024
Quality AIis available for38 Gemini languagesin preview.Quality AIsupports the following languages in addition to English:
Conversational Insights offersRule-based analysisas a preview feature to customize your conversation analyses.Rule-based analysisprovides the following customizations for your conversation analyses:
Filter conversations.
Select a percentage of your dataset.
Designate different types of analysis.
October 24, 2024
You can useQuality AIas a GA feature within the Conversational Insights console to evaluate contact center conversations and agent performance more efficiently. Quality AI automates conversation scoring so that all conversations are taken into account.
See theOverview,Basics,Setup Guide, andBest Practicespages for more details.
August 19, 2024
You can useQuality AIas a preview feature within the Conversational Insights console to evaluate contact center conversations and agent performance more efficiently. See theOverview,Basics,Setup Guide, andBest Practicespages for more details.
October 09, 2023
You can use the Conversational Insights API toingest audio conversation datain bulkfrom a Cloud Storage bucket. Optionally, you can apply redaction prior to import and transcribe the audio using custom Speech-to-Text settings.
March 31, 2023
You can use the CCAI Insights API toupload your audio conversation datafrom a Cloud Storage bucket. Optionally, you can apply redaction and analyses to your conversation prior to upload.
November 15, 2022
Topic modelingis now a GA feature.Topic modelinghelps you discover topics (call drivers) in conversations between contact center agents and end-users.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-09-04 UTC."],[[["\u003cp\u003eThis page provides production updates for Insights, including new features, bug fixes, known issues, and deprecated features.\u003c/p\u003e\n"],["\u003cp\u003eConversational Insights now offers LLM-powered topic inference as a generally available (GA) feature, which allows for real-time topic identification in conversations, but it is only available for English.\u003c/p\u003e\n"],["\u003cp\u003eQuality AI and Topic inference are both available in preview for 38 Gemini languages.\u003c/p\u003e\n"],["\u003cp\u003eRule-based analysis is available as a preview feature, allowing users to customize conversation analyses by filtering, selecting a dataset percentage, and designating different types of analysis.\u003c/p\u003e\n"],["\u003cp\u003eQuality AI is now a GA feature within Conversational Insights, enabling more efficient evaluation of contact center conversations and agent performance by automating conversation scoring.\u003c/p\u003e\n"]]],[],null,["# Conversational Insights release notes\n\nThis page documents production updates to Insights. You can\nperiodically check this page for announcements about new or updated features,\nbug fixes, known issues, and deprecated features.\n\nYou can see the latest product updates for all of Google Cloud on the\n[Google Cloud](/release-notes) page, browse and filter all release notes in the\n[Google Cloud console](https://console.cloud.google.com/release-notes),\nor programmatically access release notes in\n[BigQuery](https://console.cloud.google.com/bigquery?p=bigquery-public-data&d=google_cloud_release_notes&t=release_notes&page=table).\n\nTo get the latest product updates delivered to you, add the URL of this page to your\n[feed\nreader](https://wikipedia.org/wiki/Comparison_of_feed_aggregators), or add the\n[feed URL](https://cloud.google.com/feeds/contactcenterinsights-release-notes.xml) directly.\n\nAugust 15, 2025\n---------------\n\nQuality AI offers the virtual agent platform to help with quality management for virtual agents. The [virtual agent platform](https://cloud.google.com/contact-center/insights/docs/virtual-agent-platform) presents data for the following metrics, aggregated across all an agent's conversations:\n\n- Rate of escalation to a human agent\n- Latencies with respect to operations that an agent performs\n- Customer sentiments \nConversational Insights offers conversation **datasets** in preview. Choose from your existing history or upload new conversations to curate customized [datasets](https://cloud.google.com/contact-center/insights/docs/datasets) to test and evaluate the results of Agent Assist summarization.\n\nJuly 18, 2025\n-------------\n\nQuality AI offers **multiple scorecards** in preview. With [multiple scorecards](https://cloud.google.com/contact-center/insights/docs/qai-basics#scorecards), Quality AI can evaluate a single conversation against different criteria and provide multiple conversation scores. You can also choose which scorecard data to view on each page of the console.\n\nJune 27, 2025\n-------------\n\nQuality AI offers the [**agent engagement platform**](https://cloud.google.com/contact-center/insights/docs/agent-engagement) where contact center personnel can visualize agent performance data, including an AI-generated summary. The agent engagement platform also identifies areas where an agent excels and needs help compared to their peers. \nQualilty AI offers agent [**assessments**](https://cloud.google.com/contact-center/insights/docs/conversation-assessments) based on individual conversations. Human evaluators can also add notes to these assessments with feedback for the agent. \nQuality AI offers [**sampling rules**](https://cloud.google.com/contact-center/insights/docs/assessment-rules) that filter conversations to reduce the workload for human evaluators, saving contact centers time and money. \nAdd your own **custom tags** in [Quality AI](https://cloud.google.com/contact-center/insights/docs/qai-basics). In addition to the default Business, Compliance and Customer tags, you can apply custom tags to any question in any scorecard. \nConversational Insights offers a [devkit](https://cloud.google.com/contact-center/insights/docs/python-library-for-developers) to help developers and maintainers perform a wide range of actions, including the following:\n\n- Import one or more conversations with metadata.\n- Transcribe mono audio files.\n- Create recognizers with STT V2.\n- Transform transcript data formats from Genesys Cloud or AWS.\n- Change Conversational Insights global settings.\n\nJune 20, 2025\n-------------\n\nConversational Insights offers **sentiment analysis** in GA. [Sentiment analysis](https://cloud.google.com/contact-center/insights/docs/sentiment) determines the mood within a conversation and assigns a score: positive, neutral, or negative.\n\nMay 29, 2025\n------------\n\nConversational Insights offers a **merged analysis**, which displays the previous results of each analysis type alongside your most recent analysis result. Merged analysis eliminates the need to run every analysis multiple times.\n\nApril 28, 2025\n--------------\n\nQuality AI offers the following [conversation filters](https://cloud.google.com/contact-center/insights/docs/filtering):\n\n- CSAT\n- Sentiment score\n- Silence duration\n\nApril 23, 2025\n--------------\n\nQuality AI offers [**fine-grained access control**](https://cloud.google.com/contact-center/insights/docs/overview-of-fine-grained-access-control) in preview.\nUse [IAM custom roles](https://cloud.google.com/iam/docs/creating-custom-roles) and authorized views to control who can view which portions of your dataset.\n\nMarch 07, 2025\n--------------\n\nYou can integrate Agent Assist **summarization generators** with Conversational Insights. [Summarization](https://cloud.google.com/contact-center/insights/docs/summarization) uses existing LLM generators to automatically summarize conversations. You can then export those summaries along with your other Insights data.\n\nMarch 06, 2025\n--------------\n\nConversational Insights offers **Rule-based analysis** as a GA feature to customize your conversation analyses. [Rule-based analysis](https://cloud.google.com/contact-center/insights/docs/analysis-rule) provides the following customizations:\n\n- Filter conversations.\n- Select a percentage of your dataset.\n- Designate different types of analysis.\n\nJanuary 28, 2025\n----------------\n\n**Customer Managed Encryption Key (CMEK)** is available in GA for all Agent Assist features. [CMEK](https://cloud.google.com/contact-center/insights/docs/cmek) lets you control details of your data encryption keys, view audit logs, and control key lifecycles.\n\nDecember 23, 2024\n-----------------\n\nConversational Insights offers LLM-powered **topic inference** as a GA feature.\n[Topic inference](https://cloud.google.com/contact-center/insights/docs/topic-modeling-overview) allows you to use your topic model to analyze new conversations and identify topics in real time.\n\nThis feature is only available for English.\n\nDecember 20, 2024\n-----------------\n\n**Quality AI** is available for [38 Gemini languages](https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models#languages-gemini) in preview. [Quality AI](https://cloud.google.com/contact-center/insights/docs/qai-overview) supports the following languages in addition to English:\n\n- German\n- Italian\n- Japanese\n- Korean\n- Portuguese\n- Spanish\n- French \n[**Topic inference**](https://cloud.google.com/contact-center/insights/docs/topic-modeling-overview) is available for [38 Gemini languages](https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models#languages-gemini) in preview.\n\nNovember 14, 2024\n-----------------\n\nConversational Insights offers **Rule-based analysis** as a preview feature to customize your conversation analyses. [Rule-based analysis](https://cloud.google.com/contact-center/insights/docs/analysis-rule) provides the following customizations for your conversation analyses:\n\n- Filter conversations.\n- Select a percentage of your dataset.\n- Designate different types of analysis.\n\nOctober 24, 2024\n----------------\n\nYou can use **Quality AI** as a GA feature within the Conversational Insights console to evaluate contact center conversations and agent performance more efficiently. Quality AI automates conversation scoring so that all conversations are taken into account.\nSee the [Overview](https://cloud.google.com/contact-center/insights/docs/qai-overview), [Basics](https://cloud.google.com/contact-center/insights/docs/qai-basics), [Setup Guide](https://cloud.google.com/contact-center/insights/docs/qai-setup-guide), and [Best Practices](https://cloud.google.com/contact-center/insights/docs/qai-best-practices) pages for more details.\n\nAugust 19, 2024\n---------------\n\nYou can use **Quality AI** as a preview feature within the Conversational Insights console to evaluate contact center conversations and agent performance more efficiently. See the [Overview](https://cloud.google.com/contact-center/insights/docs/qai-overview), [Basics](https://cloud.google.com/contact-center/insights/docs/qai-basics), [Setup Guide](https://cloud.google.com/contact-center/insights/docs/qai-setup-guide), and [Best Practices](https://cloud.google.com/contact-center/insights/docs/qai-best-practices) pages for more details.\n\nOctober 09, 2023\n----------------\n\nYou can use the Conversational Insights API to [**ingest audio conversation data** in bulk](https://cloud.google.com/contact-center/insights/docs/ingest-conversations-api) from a Cloud Storage bucket. Optionally, you can apply redaction prior to import and transcribe the audio using custom Speech-to-Text settings.\n\nMarch 31, 2023\n--------------\n\nYou can use the CCAI Insights API to [**upload your audio conversation data**](https://cloud.google.com/contact-center/insights/docs/upload-audio) from a Cloud Storage bucket. Optionally, you can apply redaction and analyses to your conversation prior to upload.\n\nNovember 15, 2022\n-----------------\n\n**Topic modeling** is now a GA feature. [Topic modeling](https://cloud.google.com/contact-center/insights/docs/topic-modeling-overview) helps you discover topics (call drivers) in conversations between contact center agents and end-users.\n\nMay 20, 2022\n------------\n\nCCAI Insights now offers GA support for [**Access Transparency**](https://cloud.google.com/cloud-provider-access-management/access-transparency/docs).\n\nNovember 15, 2021\n-----------------\n\nCCAI Insights now offers GA support for VPC Service Controls integration. See the [VPCSC documentation](https://cloud.google.com/vpc-service-controls/docs) for details.\n\nOctober 12, 2021\n----------------\n\nThe CCAI Insights v1 endpoint is now GA. See the [documentation](https://cloud.google.com/contact-center/insights/docs) for details."]]
