Offline AI Voice Translator - available on the App Store

No connection. Still in conversation.

37 conversation languages with AI on your iPhone or iPad. Download models through Apple in advance and hold conversations offline.

The connection may be missing. The words need not be.

Remote areas do not always have mobile internet. Directions, accommodation questions and conversations on the road should not depend on reception. The released app recognises and translates speech directly on the device. Suitable installed iOS voices read the result aloud; without a local voice it appears as text. No cloud translation service is required.

Private conversations. On your device.

Audio and conversation text are processed only on your device and are not sent to Informatik Baums or a cloud translation service. By default, conversation text is temporary and its display is cleared when the app enters the background. The new TestFlight build 1.0.3 (33) also prepares an optional local text history with deletion and native sharing. The public Store version receives it with a corresponding update. Saving is off by default. Only “Share conversation” opens the system dialog for copying or passing text to a chosen app. Audio is not saved as a conversation recording.

Text deliberately copied into other apps remains there. Models and system voices are downloaded through Apple in advance; these downloads do not transmit conversation content.

Speak. Understand. Reply.

Select two conversation languages for both directions; swap their positions with one button. Choose the interface language independently. Tap the speaker language to start, then Stop to finish, automatically after at most 20 seconds. Original text and translation remain visible. Text input is also available; recording does not continue unnoticed in the background.

Choose the model size.

Compact, Balanced and Quality profiles require approximately 1.3 to 2.5 GB, plus installation headroom. Apple delivers packs on request. All languages share the models. Once installed, speech recognition and translation are processed on the device and local voices provide speech output.

Prepare before travelling.

After installation, setup guides you to choose and download a suitable AI model package. This requires internet access and some setup time. Allow at least 3 GB of free storage for the smallest profile; installed models use about 1.3 GB. Suitable Apple voices require additional storage. Fully install models and voices before travelling. Requires iPhone or iPad with iOS/iPadOS 26 or later and sufficient free storage. Check the App Store for device compatibility.

Automatic translations can be wrong. They do not replace qualified interpreting for safety-critical or binding decisions.

With models and voices downloaded in advance, possible uses include everyday communication in development aid or crisis areas without internet access. Have important information checked by a qualified person; the app does not replace professional emergency interpreting.

37 languages. One app.

Available on the App Store since 3 November 2026. One-time price in Germany: €4.00; no subscription or in-app purchases. Regional prices may differ; the App Store price applies. View on the App Store. Choose 37 conversation languages and 38 language/script options, including Traditional Chinese. Full menus are available for all conversation languages. The existing model packages stay the same size. Cantonese speech recognition requires Balanced or Quality. Speech output requires a suitable available Apple voice.

English, 中文(简体), हिन्दी, Español, العربية, Français, বাংলা, Português, Русский, Bahasa Indonesia, اردو, Deutsch, 日本語, मराठी, Tiếng Việt, Türkçe, 한국어, ไทย, Italiano, Bahasa Melayu, Filipino, 中文(繁體), Polski, Čeština, Nederlands, ខ្មែរ, မြန်မာ, فارسی, ગુજરાતી, తెలుగు, עברית, தமிழ், Українська, བོད་སྐད།, Қазақша, Монгол, ئۇيغۇرچە, 粵語.

Quality and available speech output differ by language and device.

Questions about the app?

App-specific support · App privacy notice · All apps

Next update: version 1.0.3

Submitted to Apple for review, not yet released: tap a translation to hear it again, clearer language selection and an optional simultaneous mode for short conversation turns. Enable the mode explicitly in Settings. It detects the spoken language locally within the selected pair; listening pauses during speech playback.

Prepare more natural voices

The app uses local Apple system voices. If speech sounds robotic, download a higher-quality voice on your iPhone:

  1. Open Settings → Accessibility → Read & Speak → Voices.
  2. Select the language and fully download an available Enhanced or Premium voice using its download icon.
  3. Return to the app and try a short spoken translation. The downloaded voice can then work offline.

Labels vary by iOS version; older versions may use “Spoken Content”. Availability and quality depend on the language and device. Selecting a preferred system voice does not pin that voice inside this app. The app prefers suitable higher-quality installed voices; it does not include its own premium voice. Without a suitable local voice, translations remain available as text.

Apple guide to Read & Speak