Model dasar pengukuran V2 perangkuman untuk chat dan suara
Tetap teratur dengan koleksi
Simpan dan kategorikan konten berdasarkan preferensi Anda.
Agent Assist kini mendukung model dasar Perangkuman V2 baru untuk data suara dan chat. Model ini memungkinkan pengguna menyesuaikan konten ringkasan dengan memilih dari bagian yang telah ditentukan berikut:
Situasi: Masalah yang memerlukan bantuan atau pertanyaan pelanggan.
Tindakan: Apa yang dilakukan agen untuk membantu pelanggan.
Penyelesaian: Hasil layanan pelanggan.
Kepuasan pelanggan: "Tidak puas" jika pelanggan tidak puas di akhir percakapan dan "Puas" jika sebaliknya.
Alasan pembatalan: Jika pelanggan meminta untuk membatalkan layanan. T/A jika tidak.
Entitas: Pasangan nilai kunci dari entitas penting yang diekstrak dari percakapan.
Fitur ini tersedia di region berikut:
us-central1, us-east1, us-west1, global.
Mengonfigurasi profil percakapan
Profil percakapan mengonfigurasi serangkaian parameter yang mengontrol saran yang diberikan kepada agen selama percakapan. Langkah-langkah berikut membuat
ConversationProfile
dengan objek
HumanAgentAssistantConfig.
Masukkan ID model Anda. Untuk menggunakan model dasar V2, jangan
tentukan ID model. Sebagai gantinya, tetapkan baseline_model_version ke 2.0.
Jika bahasa yang dimasukkan adalah bahasa Inggris, di CreateConversationProfileRequest untuk konfigurasi kueri, tentukan bagian yang akan disertakan dalam ringkasan. Bagian defaultnya adalah SITUATION, ACTION, dan RESOLUTION jika tidak ada bagian yang ditentukan.
Di CreateConversationProfileRequest untuk fitur saran, tentukan
CONVERSATION_SUMMARIZATION untuk chat atau CONVERSATION_SUMMARIZATION_VOICE untuk suara.
Contoh model dasar untuk suara
Berikut adalah contoh JSON yang menggunakan model ringkasan dasar untuk suara:
Percakapan ditangani dengan cara yang sama saat runtime, baik Anda menggunakan model dasar maupun model kustom. Lihat dokumentasi
Model kustom ringkasan
untuk mengetahui detailnya.
[[["Mudah dipahami","easyToUnderstand","thumb-up"],["Memecahkan masalah saya","solvedMyProblem","thumb-up"],["Lainnya","otherUp","thumb-up"]],[["Sulit dipahami","hardToUnderstand","thumb-down"],["Informasi atau kode contoh salah","incorrectInformationOrSampleCode","thumb-down"],["Informasi/contoh yang saya butuhkan tidak ada","missingTheInformationSamplesINeed","thumb-down"],["Masalah terjemahan","translationIssue","thumb-down"],["Lainnya","otherDown","thumb-down"]],["Terakhir diperbarui pada 2025-09-04 UTC."],[[["\u003cp\u003eAgent Assist introduces a new V2 Summarization baseline model for voice and chat, enabling customized summaries based on predefined sections.\u003c/p\u003e\n"],["\u003cp\u003eThe V2 model supports six customizable output sections: Situation, Action, Resolution, Customer Satisfaction, Reason for Cancellation, and Entities.\u003c/p\u003e\n"],["\u003cp\u003eCustomized section summaries are available only in English, and other languages will receive summaries as a single paragraph; additionally, this model is available in specific regions: us-central1, us-east1, us-west1, and global.\u003c/p\u003e\n"],["\u003cp\u003eConversation profiles, which can be set up via the console or API, allow users to configure parameters for agent suggestions and define the specific sections to be included in summaries.\u003c/p\u003e\n"],["\u003cp\u003eTo utilize the V2 baseline model, users must set the \u003ccode\u003ebaseline_model_version\u003c/code\u003e to \u003ccode\u003e2.0\u003c/code\u003e in their conversation profile configurations, and for voice or chat summaries, the respective \u003ccode\u003eCONVERSATION_SUMMARIZATION_VOICE\u003c/code\u003e or \u003ccode\u003eCONVERSATION_SUMMARIZATION\u003c/code\u003e should be specified.\u003c/p\u003e\n"]]],[],null,["# Summarization V2 baseline model for chat and voice\n\nAgent Assist now supports a new V2 Summarization baseline model for voice\nand chat data. This model enables users to customize the content of the summary by selecting from the following predefined sections:\n\n- Situation: What the customer needs help with or has question about.\n- Action: What the agent does to help the customer.\n- Resolution: Result of the customer service.\n- Customer satisfaction: \"Unsatisfied\" if the customer is unsatisfied at the end of the conversation and \"Satisfied\" otherwise.\n- Reason for cancellation: If the customer requests to cancel service. N/A otherwise.\n- Entities: The key-value pairs of important entities extracted from the conversation.\n\nThis feature is available in the following regions:\nus-central1, us-east1, us-west1, global.\n| **Note:** Summaries with customized sections are only available for the English language. For other [supported languages](/agent-assist/docs/languages) in Summarization Baseline V2, summaries are produced as a single paragraph. Low-quality summaries might get deleted and returned empty.\n\nConfigure a conversation profile\n--------------------------------\n\nA *conversation profile* configures a set of parameters that control the\nsuggestions made to an agent during a conversation. The following steps create a\n[`ConversationProfile`](/dialogflow/es/docs/reference/rest/v2/projects.conversationProfiles)\nwith a\n[`HumanAgentAssistantConfig`](/dialogflow/es/docs/reference/rest/v2/projects.conversationProfiles#ConversationProfile.HumanAgentAssistantConfig)\nobject.\n\n### Create from console\n\nWe recommend creating a conversation profile using the\n[Agent Assist console](/agent-assist/docs/conversation-profile):\n\n1. Input `Display name` for conversation profile.\n2. Select `Language` from the list.\n3. Select suggestion type `Conversation summarization` or `Conversation summarization (voice)`.\n4. Set `Suggestion model type` to `Baseline model`.\n5. Set `Baseline model version` to `2.0`.\n6. If the selected language is English, select `Output sections` to be included in the summary.\n\n### Create from API\n\n| **Note:** If using Agent Assist's [`generateStatelessSummary`](/dialogflow/es/docs/reference/rest/v2beta1/projects.locations.suggestions/generateStatelessSummary) API, this step can be skipped to instead provide the fields of the [conversation profile](/dialogflow/es/docs/reference/rest/v2/projects.conversationProfiles) within the request.\n\nTo create a conversation profile, do the following:\n\n1. Call the `create` method on the [`ConversationProfile`](/dialogflow/es/docs/reference/rest/v2/projects.conversationProfiles) resource.\n2. Provide a name for the new conversation profile.\n3. Enter your Google Cloud project ID.\n4. Enter the language code.\n5. Enter your model ID. To use the V2 baseline model, don't specify the model ID. Instead, set `baseline_model_version` to `2.0`.\n6. If entered language is English, in the `CreateConversationProfileRequest` for the query configuration, specify sections to be included in the summary. The default sections are `SITUATION`, `ACTION`, and `RESOLUTION` if no section is specified.\n7. In the `CreateConversationProfileRequest` for the suggestion feature, specify `CONVERSATION_SUMMARIZATION` for chat or `CONVERSATION_SUMMARIZATION_VOICE` for voice.\n\n#### Baseline model example for voice\n\nThe following is a JSON example that uses a baseline summarization model for voice: \n\n```\n{\n \"displayName\": \"CONVERSATION_PROFILE_NAME\",\n \"humanAgentAssistantConfig\": {\n \"humanAgentSuggestionConfig\": {\n \"featureConfigs\": [{\n \"suggestionFeature\": {\n \"type\": \"CONVERSATION_SUMMARIZATION_VOICE\"\n },\n \"queryConfig\": {\n \"sections\": {\n \"sectionTypes\": \"SITUATION\",\n \"sectionTypes\": \"ACTION\",\n \"sectionTypes\": \"RESOLUTION\",\n \"sectionTypes\": \"REASON_FOR_CANCELLATION\",\n \"sectionTypes\": \"CUSTOMER_SATISFACTION\",\n \"sectionTypes\": \"ENTITIES\"\n }\n },\n \"conversationModelConfig\": {\n \"baselineModelVersion\": \"2.0\",\n }\n }]\n }\n },\n \"languageCode\": \"en-US\"\n}\n```\n\nThe response is a `ConversationProfile` object containing the conversation\nprofile `name`: \n\n```\n{\n \"name\": \"projects/PROJECT_ID/locations/global/conversationProfiles/CONVERSATION_PROFILE_ID\",\n \"displayName\": \"CONVERSATION_PROFILE_NAME\",\n \"automatedAgentConfig\": {\n },\n \"humanAgentAssistantConfig\": {\n \"notificationConfig\": {\n },\n \"humanAgentSuggestionConfig\": {\n \"featureConfigs\": [{\n \"suggestionFeature\": {\n \"type\": \"CONVERSATION_SUMMARIZATION_VOICE\"\n },\n \"queryConfig\": {\n \"sections\": {\n \"sectionTypes\": [\n \"SITUATION\",\n \"ACTION\",\n \"RESOLUTION\",\n \"REASON_FOR_CANCELLATION\",\n \"CUSTOMER_SATISFACTION\",\n \"ENTITIES\"\n ]\n }\n },\n \"conversationModelConfig\": {\n \"baselineModelVersion\": \"2.0\",\n }\n }]\n },\n \"messageAnalysisConfig\": {\n }\n },\n \"languageCode\": \"en-US\",\n \"createTime\": \"2023-07-06T21:06:46.841816Z\",\n \"updateTime\": \"2023-07-06T21:06:46.841816Z\",\n \"projectNumber\": \"344549229138\"\n}\n```\n\n#### Baseline model example for chat\n\nThe following is a JSON example that uses a baseline summarization model for chat: \n\n```\n{\n \"displayName\": \"CONVERSATION_PROFILE_NAME\",\n \"humanAgentAssistantConfig\": {\n \"humanAgentSuggestionConfig\": {\n \"featureConfigs\": [{\n \"suggestionFeature\": {\n \"type\": \"CONVERSATION_SUMMARIZATION\"\n },\n \"queryConfig\": {\n \"sections\": {\n \"sectionTypes\": \"SITUATION\",\n \"sectionTypes\": \"ACTION\",\n \"sectionTypes\": \"RESOLUTION\",\n \"sectionTypes\": \"REASON_FOR_CANCELLATION\",\n \"sectionTypes\": \"CUSTOMER_SATISFACTION\",\n \"sectionTypes\": \"ENTITIES\"\n }\n },\n \"conversationModelConfig\": {\n \"baselineModelVersion\": \"2.0\",\n }\n }]\n }\n },\n \"languageCode\": \"en-US\"\n}\n```\n\nThe response is a `ConversationProfile` object containing the conversation\nprofile `name`: \n\n```\n{\n \"name\": \"projects/PROJECT_ID/locations/global/conversationProfiles/CONVERSATION_PROFILE_ID\",\n \"displayName\": \"CONVERSATION_PROFILE_NAME\",\n \"automatedAgentConfig\": {\n },\n \"humanAgentAssistantConfig\": {\n \"notificationConfig\": {\n },\n \"humanAgentSuggestionConfig\": {\n \"featureConfigs\": [{\n \"suggestionFeature\": {\n \"type\": \"CONVERSATION_SUMMARIZATION\"\n },\n \"queryConfig\": {\n \"sections\": {\n \"sectionTypes\": [\n \"SITUATION\",\n \"ACTION\",\n \"RESOLUTION\",\n \"REASON_FOR_CANCELLATION\",\n \"CUSTOMER_SATISFACTION\",\n \"ENTITIES\"\n ]\n }\n },\n \"conversationModelConfig\": {\n \"baselineModelVersion\": \"2.0\",\n }\n }]\n },\n \"messageAnalysisConfig\": {\n }\n },\n \"languageCode\": \"en-US\",\n \"createTime\": \"2023-07-06T21:06:46.841816Z\",\n \"updateTime\": \"2023-07-06T21:06:46.841816Z\",\n \"projectNumber\": \"344549229138\"\n}\n```\n\nHandle conversations at runtime\n-------------------------------\n\nConversations are handled the same at runtime whether you're using the baseline model or a custom model. See the\n[Summarization custom model](/agent-assist/docs/summarization#handle_conversations_at_runtime)\ndocumentation for details."]]
