將作用中且經過訓練的應用程式發佈至預備或生產端點Publish your active, trained app to a staging or production endpoint

當您完成作用中 LUIS 應用程式的建置和測試時,將它發佈至端點以供用戶端應用程式使用。When you finish building and testing your active LUIS app, make it available to your client application by publishing it to the endpoint.

警告

本文件尚未更新為最新 LUIS 入口網站的文字和螢幕擷取畫面。This document has not been updated with text and screenshots for the latest LUIS portal.

發佈Publishing

  1. 若要發佈至端點,請選取右側面板頂端的 [發佈]。To publish to the endpoint, select Publish in the top, right panel.

    右上流覽列中的 [發佈] 按鈕

  2. 選取已發佈預測端點的設定,然後選取 [發佈]。Select your settings for the published prediction endpoint, then select Publish.

    選取 [發行設定],然後選取 [發佈] 按鈕

發行位置Publishing slots

在快顯視窗顯示時選取正確的位置:Select the correct slot when the pop-up window displays:

  • 臨時.Staging.
  • 生產.Production.

藉由使用這兩個發佈位置,可讓您在已發行的端點或相同版本的兩個不同端點上,使用應用程式的兩個不同版本。By using both publishing slots, this allows you to have two different versions of your app available at the published endpoints or the same version on two different endpoints.

發行區域Publishing regions

應用程式會發佈至與 LUIS 入口網站中新增的 LUIS 預測端點資源相關聯的所有區域。The app is published to all regions associated with the LUIS prediction endpoint resources added in the LUIS portal.

例如,針對在www.luis.ai上建立的應用程式,如果您在兩個區域( westuseastus)中建立 luis 資源,並將其新增至應用程式做為資源,則會在這兩個區域中發佈應用程式。For example, for an app created on www.luis.ai, if you create a LUIS resource in two regions, westus and eastus, and add these to the app as resources, the app is published in both regions. 如需 LUIS 區域的相關資訊,請參閱區域For more information about LUIS regions, see Regions.

配置發佈設定Configuring publish settings

選取位置之後,請設定下列各內容的發佈設定:After you select the slot, configure the publish settings for:

  • 情感分析Sentiment analysis
  • 拼寫更正Spelling correction
  • 語音預備Speech priming

在您發佈之後,這些設定可從 [管理] 區段的 [發行設定] 頁面中進行審核。After you publish, these settings are available for review from the Manage section's Publish settings page. 您可以在每次發行時變更設定。You can change the settings with every publish. 如果您取消發行,則在發行期間所做的任何變更也會一併取消。If you cancel a publish, any changes you made during the publish are also canceled.

當您的應用程式發佈時When your app is published

成功發佈應用程式時,綠色成功通知會顯示在瀏覽器上方。When your app is successfully published, a green success notification appears at the top of the browser. 綠色通知列也包含端點的連結。The green notification bar also includes a link to the endpoints.

發佈連結至端點的快顯視窗

如果您需要端點 URL,請選取此連結。If you need the endpoint URL, select the link. 您也可以在頂端功能表中選取 [管理],然後選取左側功能表中的 [ Azure 資源],以取得端點 url。You can also get to the endpoint URLs by selecting Manage in the top menu, then select Azure Resources in the left menu.

情感分析Sentiment analysis

情感分析可讓 LUIS 與文字分析整合,以提供情感和關鍵片語分析。Sentiment analysis allows LUIS to integrate with Text Analytics to provide sentiment and key phrase analysis.

您不需要提供文字分析金鑰,而且不會向您的 Azure 帳戶收取此服務的費用。You do not have to provide a Text Analytics key and there is no billing charge for this service to your Azure account.

情感資料具有 1 與 0 之間的分數,指出資料的正 (較接近 1) 或負 (較接近 0) 情感。Sentiment data is a score between 1 and 0 indicating the positive (closer to 1) or negative (closer to 0) sentiment of the data. positiveneutralnegative 情感標籤是依支援的文化特性 (Culture) 而定。The sentiment label of positive, neutral, and negative is per supported culture. 目前,只有英文支援情感標籤。Currently, only English supports sentiment labels.

如需使用情感分析之 JSON 端點回應的詳細資訊,請參閱情感分析For more information about the JSON endpoint response with sentiment analysis, see Sentiment analysis

拼寫更正Spelling correction

更正拼寫是在 LUIS 使用者語句預測之前進行。Corrections to spelling are made before the LUIS user utterance prediction. 在回應中,您可以看到原始語句的任何變更(包括拼寫)。You can see any changes to the original utterance, including spelling, in the response.

語音預備Speech priming

語音預備是在將文字轉換成語音之前,使用將 LUIS 模型傳送至語音服務的程式。Speech priming is the process of using sending the LUIS model to Speech services prior to conversion of text to speech. 這可讓語音服務更精確地為您的模型提供語音轉換。This allows the speech service to provide speech conversion more accurately for your model. 這可讓 bot 語音和 LUIS 要求和回應呼叫一次,方法是進行一次語音通話並取回 LUIS 回應。This allows bot Speech and LUIS requests and responses in one call by making one speech call and getting back a LUIS response. 它提供的整體延遲較低。It provides less latency overall.

後續步驟Next steps

  • 請參閱管理金鑰,將金鑰新增至 LUIS 的 Azure 訂用帳戶金鑰,以及了解如何設定 Bing 拼字檢查金鑰並且在結果中包含所有意圖。See Manage keys to add keys to Azure subscription key to LUIS and how to set the Bing Spell Check key and include all intents in results.
  • 如需如何在測試主控台中測試已發佈應用程式的指示,請參閱訓練和測試應用程式See Train and test your app for instructions on how to test your published app in the test console.