VoiceTrans+

Fully offline AI voice interpreter. Just speak and it translates both ways.

A fully on-device AI voice interpreter. Once you download the large AI model (Gemma 4 E4B) to your smartphone, the app works without any internet connection. Just press the button and speak: it instantly translates and reads aloud in both directions.

■ Works offline / keeps your data on the device After the model is downloaded, normal voice interpreting runs completely offline. You can use it even where there is no signal, and your conversations are never sent to the internet (everything is processed on the device).

■ Important notice before you start (please read)

• On the first launch only, the app downloads the AI model (about 3.7 GB) over the network. Wi-Fi is strongly recommended (this can take several to a dozen-plus minutes depending on your connection speed).
• On the first launch, the app builds a model cache (optimized data) tailored to your device, which takes about 2-3 minutes. Please do not close the app while "Loading model..." is shown (closing it midway means starting over next time).
• On every launch after that, loading the model still takes about 6-7 seconds.

■ Main features

• Just press and speak: press the Language 1 / Language 2 button and speak, and it translates into the other person's language and reads it aloud.
• The first language is shown automatically based on your device language. You can change it among 19 languages from the main screen.
• Word registration: you can register your own words, or translations for specified words.
• Online dictionary link: using Claude / Gemini / OpenAI, you can register words translated into each language (optional feature, first-use consent required).
• Scene packs: switch terminology to fit the scene, such as travel or medical conversation. By also using Claude / Gemini / OpenAI to check the conversation, you can get more natural, scene-specific translations.

■ Subscription VoiceTrans+ offers two auto-renewing annual plans, each starting with a 7-day free trial:

• General: voice interpreting in all 19 languages.
• Pro: everything in General, plus specialized packs such as medical terminology. After the free trial, the plan auto-renews every year at the price shown on the App Store unless you cancel at least 24 hours before it ends. Manage or cancel anytime in your App Store account settings.

■ Privacy Normal voice interpreting is entirely processed on the device and is not sent anywhere. Only when you choose "Consult online AI" is the selected text sent to the AI provider you yourself choose (Claude / Gemini / OpenAI) (first-use opt-in consent required).

• The wording of the disclaimer, consent, and privacy policy will be reviewed by experts before release.

Terms of Use (EULA): https://www.apple.com/legal/internet-services/itunes/dev/stdeula/