Google brings 'AI Edge Eloquent' app for offline speech-to-text on iOS - Business Standard

2026-04-07 · Source: artifical intelligence via Google News · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation · Depth: Fundamental Awareness, quick

Summary

Google has released AI Edge Eloquent, a new iOS dictation application designed for real-time, offline speech-to-text conversion. The app utilizes Google's Gemma AI models for automatic speech recognition, enabling it to function without an internet connection after initial model downloads. A key feature is its ability to automatically remove filler words like "um" and "uh" and correct mid-sentence stumbles, producing clean, professional text. Users can opt for on-device processing or enable a cloud mode that leverages Gemini models for enhanced text refinement and can integrate keywords from Gmail. The app also offers text formatting options, custom word additions, and transcription history tracking, and is compatible with iOS 16+, macOS 13+ (Apple silicon), and visionOS 1.0+.

Key takeaway

For professionals who frequently dictate notes or drafts, Google AI Edge Eloquent offers a robust solution for generating clean, edited text directly from speech, even offline. Your workflow can become more efficient by leveraging its AI-driven cleanup of filler words and stumbles. Consider integrating this app to streamline documentation and content creation, especially when internet access is unreliable.

Key insights

Google's AI Edge Eloquent offers real-time, offline speech-to-text with AI-driven filler word removal.

Principles

• On-device AI enhances privacy and accessibility.
• AI can refine natural speech into professional text.

Method

The app converts speech to text using Gemma AI, then refines it by removing filler words and stumbles, optionally using Gemini models for further enhancement.

In practice

• Use custom words for specialized jargon.
• Switch to cloud mode for Gemini-powered refinement.

Topics

• Google AI Edge Eloquent
• Offline Speech-to-Text
• Gemma AI
• Filler Word Removal
• iOS Dictation

Best for: NLP Engineer, AI Product Manager, Entrepreneur, General Interest, Tech Journalist

Related on AIssential

• TurnGuide: Enhancing Meaningful Full Duplex Spoken Interactions via Dynamic Turn-Level Text-Speech Interleaving — TurnGuide enhances full-duplex SLMs by planning turn-level text guidance before speech, improving semantic coherence and natural flow.
• The best AI dictation apps, tested and ranked — Modern AI dictation apps offer enhanced accuracy, privacy, and customization through advanced LLMs and speech-to-text models.
• TAI #198: Real-Time Speech AI Gets Serious: Google and OpenAI Race to Own the Voice Layer — Real-time speech AI is maturing into deployable infrastructure with significant advancements in quality, speed, and cost efficiency.
• Do You Actually Need to Pay for Transcription Software? — Advanced AI transcription and post-processing can be achieved without expensive subscriptions by leveraging free or existing tools.
• Nothing introduces an AI-powered dictation tool — Essential Voice offers AI-powered dictation with filler word removal, custom shortcuts, and multi-language translation.
• From Text to Voice: A Reproducible and Verifiable Framework for Evaluating Tool Calling LLM Agents — Converting text-based tool-calling benchmarks to audio reveals significant model- and task-dependent performance gaps.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by artifical intelligence via Google News.