Google AI Edge Eloquent: Offline AI Dictation on iOS
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisis ton rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
Google Introduces an Offline AI Dictation App on iOS
Google has recently launched, without much public announcement, a new voice dictation app called Google AI Edge Eloquent. This app, currently available only on iOS, positions itself as a direct competitor to existing solutions such as Wispr Flow, SuperWhisper, and Willow. The app stands out for its ability to function offline, a feature that could appeal to many users concerned about privacy or data consumption.
The app is available for free download, and its operation relies on automatic speech recognition (ASR) models developed around Gemma technology. Once these models are installed on the device, users can start dictating directly on their phones. The app offers real-time transcription, and when a pause is detected, it automatically cleans up the text by removing filler words such as “um” and “ah.”
Advanced Features for Optimized Transcription
Beneath the transcription, the app offers several text formatting options such as “Key Points,” “Formal,” “Short,” and “Long.” These options allow users to transform the dictated text according to their specific needs. Additionally, the app provides the option to disable cloud mode, enabling data processing to occur solely locally. This means that when cloud mode is activated, the app uses cloud-based Gemini models to refine the text.
The app also allows users to import certain keywords, names, and specific jargon from their Gmail account if they wish. It is also possible to add custom words to the list, thus offering an enhanced personalization of the user experience.
Comprehensive Management of Transcription Sessions
Google AI Edge Eloquent keeps a history of transcription sessions, allowing users to search through them. The app provides detailed information about the words dictated during the last session, the dictation speed in words per minute, and the total number of words spoken. This feature is particularly useful for users who want to track and analyze their use of the app.
The app's description in the App Store emphasizes that “Google AI Edge Eloquent is an advanced dictation app designed to bridge the gap between natural speech and professional-ready text.” Unlike traditional dictation software, this app uses artificial intelligence to capture the user's intent, automatically eliminating hesitations and corrections mid-sentence to produce clear and precise text.
Towards Availability on Android
Although the app is currently limited to iOS, the description mentions a future Android version. This version promises a “seamless Android integration,” allowing the app to become the default keyboard for users, providing universal access across all text fields. Furthermore, the app could incorporate a floating button feature, similar to that of Wispr Flow, to facilitate access to transcription from any screen.
AI-based transcription apps are experiencing growing popularity, thanks to the continuous improvement of speech-to-text conversion models. With the launch of this experimental app, Google is aligning itself with this trend. If this test proves successful, it is likely that enhanced transcription features will also be deployed on Android, thereby expanding the accessibility and efficiency of these tools to an even broader audience.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.