Stay organized with collections
Save and categorize content based on your preferences.
Overview
In order to make suggestions, Agent Assist relies on data in the form of
either documents or conversation transcripts. You must upload your data
before you can use Agent Assist. The linked tutorials at the bottom of
this page walk you through the steps required to upload your data using the
Agent Assist console.
You can use the console to configure Agent Assist features and test out
how they function.
Data types
Agent Assist uses two types of data to make suggestions to human agents:
Conversation datasets, which are collections of conversation transcripts,
and knowledge bases, which are collections of knowledge documents
(articles or FAQ documents). Agent Assist features analyze a conversation
in real time and make suggestions to human agents based on either conversation
datasets or knowledge bases.
Smart Reply and
Summarization surface suggestions
trained on conversation datasets. Smart Reply suggests text responses to agents
as they converse with an end-user, and Summarization suggests conversation
summaries after an exchange with an end-user has completed. Each model is custom
by definition because each conversation dataset is made up of your own
conversation transcript data.
The FAQ Assist and
Article Suggestion features draw on
knowledge bases to make recommendations instead of conversation datsets. Article
Suggestion suggests knowledge documents (such as articles) to agents during a
conversation. FAQ Assist makes suggestions based on FAQ pairs (an FAQ question
and its associated answer) rather than entire articles. You do not need
to train a custom model in order to use these features: Agent Assist
uses default baseline suggestion models to make suggestions from your
knowledge base. If you want to upload your own conversation data to train a
custom suggestion model for Article Suggestion, please contact your Google
representative. Custom suggestion models are not available for FAQ Assist.
[[["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 2025-08-28 UTC."],[[["\u003cp\u003eAgent Assist uses either conversation transcripts or knowledge documents to provide suggestions, requiring data upload before use.\u003c/p\u003e\n"],["\u003cp\u003eConversation datasets, used for features like Smart Reply and Summarization, are collections of conversation transcripts that train custom suggestion models.\u003c/p\u003e\n"],["\u003cp\u003eKnowledge bases, used for FAQ Assist and Article Suggestion, are collections of knowledge documents that use baseline suggestion models, without requiring custom model training.\u003c/p\u003e\n"],["\u003cp\u003eThe Agent Assist console is utilized for design-time and model testing, while all live operations must be executed via the API.\u003c/p\u003e\n"],["\u003cp\u003eUsers can begin by creating a conversation dataset or a knowledge base to start using Agent Assist features.\u003c/p\u003e\n"]]],[],null,["# Upload data\n\nOverview\n--------\n\nIn order to make suggestions, Agent Assist relies on data in the form of\neither documents or conversation transcripts. You must upload your data\nbefore you can use Agent Assist. The linked tutorials at the bottom of\nthis page walk you through the steps required to upload your data using the\n[Agent Assist console](https://agentassist.cloud.google.com).\nYou can use the console to configure Agent Assist features and test out\nhow they function.\n| **Note:** The Agent Assist console can be used during design-time and model testing phases only. All runtime operations must call the API directly. You can use the API to carry out the same actions described in the [tutorials](/agent-assist/docs/tutorials), which focus on using the console. See the Agent Assist [how-to guides](/agent-assist/docs/how-to) for details on sending API requests.\n\nData types\n----------\n\nAgent Assist uses two types of data to make suggestions to human agents:\n**Conversation datasets** , which are collections of conversation transcripts,\nand **knowledge bases** , which are collections of *knowledge documents*\n(articles or FAQ documents). Agent Assist features analyze a conversation\nin real time and make suggestions to human agents based on either conversation\ndatasets or knowledge bases.\n\n[Smart Reply](/agent-assist/docs/smart-reply) and\n[Summarization](/agent-assist/docs/summarization-console) surface suggestions\ntrained on conversation datasets. Smart Reply suggests text responses to agents\nas they converse with an end-user, and Summarization suggests conversation\nsummaries after an exchange with an end-user has completed. Each model is custom\nby definition because each conversation dataset is made up of your own\nconversation transcript data.\n\nThe [FAQ Assist](/agent-assist/docs/faq) and\n[Article Suggestion](/agent-assist/docs/article-suggestion) features draw on\nknowledge bases to make recommendations instead of conversation datsets. Article\nSuggestion suggests *knowledge documents* (such as articles) to agents during a\nconversation. FAQ Assist makes suggestions based on FAQ pairs (an FAQ question\nand its associated answer) rather than entire articles. You do not need\nto train a custom model in order to use these features: Agent Assist\nuses default baseline suggestion models to make suggestions from your\nknowledge base. If you want to upload your own conversation data to train a\ncustom suggestion model for Article Suggestion, please contact your Google\nrepresentative. Custom suggestion models are not available for FAQ Assist.\n\nWhat's next\n-----------\n\nCreate a [conversation dataset](/agent-assist/docs/conversation-dataset) or a\n[knowledge base](/agent-assist/docs/knowledge-base)."]]
