Google has quietly expanded its footprint in the mobile productivity space with the release of Google AI Edge Eloquent, a new offline-first AI dictation app now available on iOS. Launched on Monday, April 6, 2026, the application aims to transform the often-cluttered process of speech-to-text into a streamlined workflow for professional-grade prose.
Unlike traditional dictation tools that record every verbal stumble, Google AI Edge Eloquent leverages lightweight Gemma AI models to process speech directly on the device. This approach allows users to dictate their thoughts naturally, while the AI works in the background to filter out the noise of spontaneous speech, positioning Google to compete directly with existing AI-driven transcription services like Wispr Flow, SuperWhisper, and Willow via TechCrunch.
The app is available as a free download on the iOS App Store. Upon installation, users must download the Gemma-based automatic speech recognition (ASR) models to enable the app’s core functionality. Once these models are local, the app provides real-time transcription, offering a seamless bridge between spoken ideas and polished text.
Beyond Simple Transcription: Polishing Natural Speech
The primary differentiator for Google AI Edge Eloquent is its ability to distinguish between a user’s intended meaning and the artifacts of natural conversation. Standard dictation software typically transcribes every “um,” “ah,” and mid-sentence correction verbatim, often leaving the user with a messy transcript that requires significant manual editing.
Eloquent addresses this by automatically filtering out filler words and polishing the resulting text the moment the user hits pause. According to the company’s App Store description, “Google AI Edge Eloquent is an advanced dictation app engineered to bridge the gap between natural speech and professional, ready-to-use text. Unlike standard dictation software that transcribes stumbles and filler words verbatim, Eloquent utilizes AI to capture your intended meaning. It automatically edits out ‘ums,’ ‘uhs,’ and mid-sentence self-corrections, outputting clean, accurate prose” via TechCrunch.
To further refine the output, the app provides several transformation options located below the transcript. Users can instantly modify the tone or length of their text by selecting options such as:
- Key points: Summarizes the dictation into a concise list.
- Formal: Adjusts the language for professional correspondence.
- Short: Condenses the text for brevity.
- Long: Expands the text for more detail.
Local Processing vs. Cloud-Based AI
A central feature of the app is its “offline-first” architecture. By running Gemma language models directly on the device, Google allows for local-only processing, which is critical for users concerned with privacy or those working in environments without reliable internet connectivity via Startup Fortune.
However, the app does offer a hybrid approach. Users can enable a “cloud mode,” which shifts the text cleanup process from the local Gemma models to Google’s more powerful cloud-based Gemini models. This allows users to choose between the speed and privacy of on-device processing and the potentially higher linguistic sophistication of cloud AI.
Personalization and Productivity Metrics
To ensure accuracy with specialized terminology, Google AI Edge Eloquent includes several customization features. The app can import specific keywords, names, and industry jargon directly from a user’s Gmail account, reducing the likelihood of transcription errors for personalized or professional terms. Users have the ability to manually add their own custom words to the app’s recognition list.
For those tracking their productivity, the app maintains a detailed history of transcription sessions. Users can search through previous sessions and view specific performance metrics, including:
- Words dictated in the most recent session.
- The user’s word-per-minute (WPM) speaking speed.
- The total number of words spoken across the session.
Market Impact and Availability
The release of Google AI Edge Eloquent marks a strategic move into the high-end speech-to-text market, challenging niche players like Willow and SuperWhisper via AI Chief. By offering the tool for free and leveraging its existing ecosystem (such as Gmail integration), Google is lowering the barrier to entry for AI-powered dictation.
While the app is currently only available for iOS users, a broader rollout is planned. The App Store description explicitly references a forthcoming Android version, suggesting that Google intends to bring these offline AI capabilities to its own mobile operating system in the near future.
Key Takeaways: Google AI Edge Eloquent
| Feature | Detail |
|---|---|
| Platform | iOS (Android version referenced) |
| AI Models | Gemma (Local) / Gemini (Cloud) |
| Core Benefit | Offline-first, automatic filler word removal |
| Cost | Free to download |
| Customization | Gmail import and custom word lists |
As AI continues to move toward “edge” computing—where processing happens on the device rather than a remote server—tools like Eloquent represent a shift toward faster, more private user experiences. The industry will be watching to see how quickly the Android version arrives and how the competitive landscape shifts in response to Google’s entry.
We will continue to monitor the App Store for updates regarding the Android release and any further feature expansions. Do you use AI dictation in your professional workflow? Share your thoughts in the comments below.