Google has launched a new dictation application for iOS devices named Google AI Edge Eloquent. This app operates primarily offline, allowing users to transcribe speech without an internet connection after downloading its models. It is available for free on the App Store.
The core functionality relies on Gemma-based automatic speech recognition systems. Once these models are installed locally, the app begins live transcription as you speak. It automatically filters out filler words such as “um” and “ah” when you pause dictation, producing polished text.
Below the transcript, users can select from several text transformation options. These include “Key points,” “Formal,” “Short,” and “Long” to modify the output. The app also features a cloud mode toggle; when enabled, it uses Gemini models hosted online for enhanced text cleanup, while disabling it restricts processing to local resources only.
Google AI Edge Eloquent can import specific keywords, names, and technical jargon from a user’s Gmail account if permission is granted. Additionally, it supports adding custom words to a personal dictionary to improve accuracy for specialized terms.
The application maintains a history of all transcription sessions, which are searchable. It displays metrics like words dictated in the last session, words-per-minute speed, and the total word count spoken over time.
According to the 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.”
While currently exclusive to iOS, the App Store listing previously mentioned an Android version, noting “seamless Android integration” that would allow setting it as the default keyboard for system-wide access. It also referenced a floating button feature similar to Wispr Flow’s implementation on Android for easy transcription access from any screen. However, Google has since updated the listing to remove references to the Android app, adding instead that an iOS keyboard is coming soon.
This release positions Google against competitors like Wispr Flow, SuperWhisper, and Willow in the growing market for AI-powered transcription tools. As speech-to-text models advance, such apps are gaining popularity among users. With this experimental offering, Google is entering this trend, and if successful, it could lead to improved transcription features being integrated into Android in the future.
