Google Launches New Offline Voice Dictation App for iPhone

AI-powered app turns speech into polished text without internet connection

Google has launched a new voice dictation app for iPhone called Google AI Edge Eloquent, offering users a smarter way to convert speech into text — even without an internet connection.

The app is now available for free on the Apple App Store, and it is designed to make speech-to-text faster, cleaner, and more useful for everyday writing tasks. With this move, Google is expanding its presence in the growing market of AI productivity tools for mobile devices.

Built for real-time speech-to-text on the device

Google AI Edge Eloquent uses automatic speech recognition (ASR) technology powered by Google’s Gemma AI models. As a result, users can speak naturally and see their words appear as text in real time directly on the iPhone.

However, the app does more than basic transcription. Instead of simply copying spoken words exactly as they are, it also improves the output by removing filler words such as “um” and “uh”, then refining the text into a more polished version.

This makes it especially useful for notes, messages, drafts, work tasks, and content creation.

Smart text tools for better writing

One of the app’s biggest advantages is that it does not stop at transcription. After dictation, users can also choose different AI-powered text options, including:

• Summarize key points
• Make the text more formal
• Shorten the content
• Expand the writing
• Refine the phrasing

Because of that, the app can help turn casual speech into text that feels more professional and ready to use.

At the moment, the app supports English only, but Google may expand language support in future updates.

Works offline for speed and privacy

A major feature of Google AI Edge Eloquent is its ability to run fully on-device, meaning it can work without internet access.

This gives users two important benefits:
first, better privacy, since voice processing can stay on the phone; and second, faster response times, especially when there is no strong connection available.

On the other hand, users can also enable cloud-based processing when connected online. In that case, the app uses Google’s Gemini models to further improve and optimize the text.

So, in practice, users get the flexibility of both offline convenience and cloud-enhanced AI editing.

Extra features for daily productivity

Google has also added a few useful features to make the app more practical for regular use.

For example, the app can:

• Keep a history of past dictation sessions
• Let users search inside saved transcripts
• Show usage insights such as:
  • Words per minute
  • Total spoken word count

In addition, it can support custom names and terms, and in some cases pull personalized vocabulary from Gmail, while also allowing manual additions.

These features can be especially helpful for users who often dictate emails, ideas, work notes, or repeated terminology.

How it differs from traditional dictation apps

According to Google, the goal of the app is not just to convert speech into text, but to transform natural spoken language into usable written content.

That is a key difference from many standard dictation tools, which often reproduce speech exactly as spoken — including pauses, repeated words, corrections, and verbal filler.

By contrast, Google AI Edge Eloquent is built to make spoken input feel closer to a clean first draft, which could make it more attractive for people who want to save time while writing on mobile.

iPhone first, Android next

For now, Google AI Edge Eloquent is available only on iPhone (iOS). However, an Android version is reportedly in development and is expected to include deeper system-level integration.

That future version may offer additional features such as:

• Keyboard integration
• Floating dictation access
• Wider system support

If Google expands the app successfully, it could become one of the strongest AI dictation tools available on smartphones.

Conclusion

Overall, Google AI Edge Eloquent is more than just another voice typing app. It combines offline speech recognition, AI-powered editing, real-time transcription, and privacy-focused processing into one tool designed for modern mobile productivity.

Although it currently supports only English and is limited to iPhone, it already shows strong potential — especially for users who want faster, smarter, and cleaner voice-to-text writing.