Google has quietly released a new offline-first AI dictation app for iOS called "Google AI Edge Eloquent," entering a growing market for speech-to-text tools that operate directly on devices. The free app allows users to dictate speech on their phones after downloading buit-in automatic speech recognition (ASR) models based on Google's Gemma AI system. During use, the application displays a live transcript and automatically refines the text when the user pauses, removing filler words such as "um' and "ah" while polishing the output.
Users can also transform their dictation through built-in formatting options, including "Key points" "Formal," "short," and Long."
Google says the app can operate entirely offline. When cloud mode is disabled, processing occurs locally on the device. If cloud mode is enabled, the app uses cloud-based Gemini models to further refine text.
The application can also import specific keywords, names, and jargon from a user's Gmail account and allows additional custom terms to be added manually.
A history feature lets users review past dictation sessions and search transcripts. The app also provides usage metrics, including words spoken in the most recent session, speaking speed in words per minute, and the total number of words dictated.
According to its 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.”
Although the app is currently available only on iOS, the App Store listing also referenced Android compatibility before being updated later Monday. The revised listing removed mentions of the Android app but noted that an iOS keyboard version is “coming soon.”
The earlier description said the Android version would offer “seamless Android integration,” allowing users to set the app as their default keyboard across the operating system. It also referenced a floating transcription button similar to the one used by voice-dictation app Wispr Flow.
AI-powered transcription tools have gained traction as speech-recognition models improve, enabling more accurate real-time dictation. With the experimental launch of Eloquent, Google appears to be testing the technology ahead of potentially integrating similar features across its broader Android ecosystem.
