2)Gemini はまず、システム プロンプトに基づいて Markdown 形式で(AI)旅行プランを作成します。次に、その結果を再度処理して、旅行プランに記載されている場所を抽出します。Google Cloud Platform(Maps)の Geo-Coding API と Places API を使用して、旅行プランに記載されている場所に対応するマーカーをインタラクティブ マップに動的に表示します。これにより、非常に完成度の高いエクスペリエンスが実現します。
構成
Android
Firebase
Geo-Coding API
Google Cloud Platform の Places API
チーム
By
Geolingo
差出人
シンガポール
[[["わかりやすい","easyToUnderstand","thumb-up"],["問題の解決に役立った","solvedMyProblem","thumb-up"],["その他","otherUp","thumb-up"]],[["必要な情報がない","missingTheInformationINeed","thumb-down"],["複雑すぎる / 手順が多すぎる","tooComplicatedTooManySteps","thumb-down"],["最新ではない","outOfDate","thumb-down"],["翻訳に関する問題","translationIssue","thumb-down"],["サンプル / コードに問題がある","samplesCodeIssue","thumb-down"],["その他","otherDown","thumb-down"]],[],[],[],null,["# Geolingo\n\n[More Apps](/competition/vote) \n\nGeolingo\n========\n\nAI generated language flashcards made from YOUR memories \nVote \nVoted!\nWhat it does\n\nOur app is an Engaging AI generated language flashcard maker that makes language flashcards from images the user uploads, powered by Gemini. User creates a new card deck, adds in images, and Gemini takes over to extract text, and translate the extracted text, hence building an image and its translated word pair. User can create or join parties, collaborating with shared card decks. User can view global card decks contributed by Geolingo's community by selecting the markers shown on the map. User can also use Geolingo to plan his trip itinerary, and an interactive map shows user visually where the locations are, including important amenities like Clinic, Gym, Restaurants near the entered destination. \n\n1) Gemini Flash's Multimodal capabilities was used to first extract the texts in the image given regardless of whatever language it was. Gemini Flash was then used again to translate the extracted text, forming an image-word pair that's stored in Firestore. The whole app was only made possible because of the NoSQL format firebase Firestore has. \n\n2) Gemini will first build the (AI)tinerary using Markdown format, guided by system prompts, and then process the result again to extract the places mentioned in the itinerary. We use Geo-Coding and Places API from Google Cloud Platform(Maps) to dynamically display the markers on the interactive map, corresponding to the locations mentioned in the itinerary. This makes a very complete experience. \nBuilt with\n\n- Android\n- Firebase\n- Geo-Coding API\n- Places API from Google Cloud Platform \nTeam \nBy\n\nGeolingo \nFrom\n\nSingapore \n[](/competition/vote)"]]