Google Enters AI-Powered Transcription Market with Offline Capabilities

Google has quietly released an AI dictation app called Google AI Edge Eloquent, which works offline and offers advanced transcription features.

The app, currently available on iOS, uses Gemma-based automatic speech recognition (ASR) models to transcribe spoken words into text. Users can see the live transcription and pause the app to filter out filler words like "um" and "ah." The app also offers options to transform the text into different formats, such as key points, formal, short, and long.

One of the standout features of Google AI Edge Eloquent is its ability to use local-only processing, allowing users to work offline without relying on cloud-based services. The app can also import keywords, names, and jargon from the user's Gmail account and add custom words to the list.

The app displays a history of transcription sessions, allowing users to search through previous sessions and view their words-per-minute speed and total number of words spoken. Google's App Store description highlights the app's ability to capture the user's intended meaning, editing out stumbles and filler words to produce clean, accurate prose.

While the app is currently only available on iOS, the App Store description mentions an upcoming Android version that will offer seamless integration and a floating button feature for easy access to transcription.

Google's entry into the AI-powered transcription market is notable, as the company joins a trend of users seeking better speech-to-text models. If successful, this experimental app could lead to improved transcription features across Android devices.

Key Features of Google AI Edge Eloquent

• Advanced automatic speech recognition (ASR) models
• Transcription filtering and customization options
• Local-only processing for offline use
• Integration with Gmail account for keyword, name, and jargon import
• Custom word list
• Transcription history and session search

Implications of Google's AI Dictation App

• Improved transcription features across Android devices
• Increased competition in the AI-powered transcription market
• Potential for better speech-to-text models in the future

What's Next for Google AI Edge Eloquent

• Release of the Android version
• Further development and refinement of the app's features
• Potential expansion to other platforms, such as web or desktop applications