最後更新:2026 年 9 月 7 日
azVoxtrans(下稱「本 App」)是一款語音轉文字工具(翻譯功能開發中),由 iPost 開發。 我們把隱私做得很簡單:本 App 不收集、不儲存、不傳送你的任何個人資料。 沒有帳號、沒有後端伺服器、沒有分析追蹤、沒有廣告。
本 App 需要麥克風權限,才能將你的語音轉換成文字。錄音只在你按下錄音鍵之後才開始, 停止即結束,不會在背景聆聽。
語音辨識在你的裝置上完成,本 App 不會把你的語音送到我們的伺服器 ——我們沒有伺服器。實際辨識由作業系統內建的語音辨識服務執行 (iOS 的 Speech 框架、Android 的語音辨識服務)。在極少數情況下, 若裝置尚未下載對應語言的離線辨識模型,系統可能改由平台自身的線上服務處理; 該行為由 Apple 或 Google 執行,適用其各自的隱私權政策。你可在裝置設定中 下載離線語言模型以確保全程在裝置上進行。
翻譯功能目前仍在開發中,尚未開放使用,因此本 App 現階段不會處理或傳送任何翻譯內容。 該功能推出時將採用裝置端的離線翻譯,不把原文或譯文送到伺服器; 屆時我們會更新本頁說明。
錄下的音檔與轉出的文字,都存放在你裝置上的本 App 專屬空間中。 這些內容只有你能存取,我們接觸不到,也不會上傳到任何地方。 刪除檔案或移除 App 即完全清除。
你導入的音檔、影片或文字檔,僅在你選擇時才被讀取,用途限於播放、轉寫或編輯, 同樣只留在你的裝置上。本 App 不會瀏覽你未選擇的檔案。
當你匯出音檔或文字時,是透過系統內建的檔案/分享功能, 由你自己決定儲存位置或送出對象。本 App 不經手、不保留這些內容的副本。
本 App 的核心功能不需要網路。會用到連線的情況僅有兩種: 下載離線語言辨識模型(由作業系統或平台提供),以及 內購與恢復購買(由 Apple App Store 或 Google Play 處理)。 本 App 僅向平台查詢「你是否已擁有該商品」的授權狀態, 收不到你的姓名、付款方式或交易明細。 相關資料的處理適用 Apple 與 Google 各自的隱私權政策。
本 App 僅要求麥克風與語音辨識權限,用途如上所述; 以及在你主動導入或匯出檔案時,存取你所選擇的檔案。 不存取通訊錄、定位、相機、照片或其他資料。你可隨時在系統設定中撤銷這些權限, 撤銷後相關功能將無法使用,其餘功能不受影響。
本 App 不特別針對兒童設計,且不論使用者年齡都不收集任何個人資料。
本 App 不內嵌第三方分析或廣告 SDK,也不會把資料分享給第三方。
若本政策有更新,我們會修改本頁上方的「最後更新」日期。
有任何隱私相關問題,請來信 support@ipost.tw。
Last updated: September 7, 2026
azVoxtrans (the “App”) is a speech-to-text tool by iPost (translation is in development). The App does not collect, store, or transmit any personal data. There are no accounts, no backend servers, no analytics, and no ads.
Microphone and speech recognition. The App needs microphone access to convert your speech to text. Recording starts only when you tap record and stops when you stop it; the App never listens in the background. Recognition runs on your device — your audio is never sent to our servers, because we have none. Recognition itself is performed by the operating system's built-in service (Apple's Speech framework on iOS, the system speech recognizer on Android). In rare cases where your device has not downloaded the offline model for a language, the platform may process it via its own online service; that is carried out by Apple or Google under their respective privacy policies. You can download offline language models in your device settings to keep everything local.
Translation is still in development and not yet available, so the App currently processes no translation content. When it ships it will run on-device, with neither source text nor translation sent to any server; this page will be updated then.
Your recordings and text are stored in the App's own space on your device. Only you can access them; we cannot, and they are never uploaded. Deleting the files or the App removes them completely. Imported files are read only when you choose them and stay on your device. Exporting or sharing uses the system file and share features — you decide the destination and we keep no copy.
Network access. Core features work entirely offline. Networking is used only to download offline speech models, and for in-app purchase and restore, handled by the Apple App Store or Google Play; the App only learns whether you own the product and never sees your name, payment method, or transaction details.
Permissions are limited to microphone and speech recognition, plus access to files you explicitly import or export. The App does not access contacts, location, camera, or photos. It is not directed at children and collects no personal data regardless of age. No third-party analytics or ad SDKs are embedded.
Questions: support@ipost.tw.