Skip to content
Trending storyDeveloping

Google adds on-device voice chat with Gemma to AI Edge Gallery

2 articles2 sourcessince Nov 9Last article 2h ago ·

Overview

AISummary of 2 articles

Google says AI Edge Gallery's AI Chat now lets users speak to Gemma models, with dictation streamed into the prompt box and responses read aloud with sentence highlighting.

Google's Gemma account on X says the speech pipeline runs offline on the device.

Google Developers Blog adds that the on-device speech recognition comes from ML Kit and covers 100+ languages at 10x real-time speed, and that text-to-speech starts speaking in about 100 ms. The blog says the feature pairs with small models such as Gemini Nano and Gemma 4, and that open-weight speech bundles for MediaPipe Tasks and LiteRT plus open-source reference code are available. The blog says users can try the voice features by downloading the latest AI Edge Gallery app on Android or iOS.

Written by AI from the articles below · updated Nov 10, 1:05 PM ET

Check the sources:

Latest developmentNov 10, 12:30 PM ET
Google adds offline voice chat with Gemma to AI Edge Gallery

Developments

2 developments

  1. Nov 10, 12:30 PM ET · 1 article
    Google adds offline voice interaction to AI Edge Gallery chat
    Google Gemma: Google adds offline voice chat with Gemma to AI Edge Gallery
  2. Nov 9, 7:00 PM ET · 1 article
    Google adds on-device speech recognition and text-to-speech to AI Edge Gallery
    Google Developers Blog: Google adds on-device speech recognition and text-to-speech to AI Edge Gallery

Article timeline

The articles in this story. Times are ET.

Nov 10
  1. Google GemmaOfficial
    Google adds offline voice chat with Gemma to AI Edge Gallery

    AIGoogle says AI Edge Gallery's main AI Chat now lets users talk directly with Gemma models. The new on-device speech recognition and text-to-speech pipeline streams live voice dictation into the prompt box and reads responses aloud with real-time sentence highlighting, running offline on the device.

    Video from @googlegemma's post
Nov 9
  1. Google Developers BlogOfficialPick
    Google adds on-device speech recognition and text-to-speech to AI Edge Gallery

    AIGoogle introduces on-device automatic speech recognition and text-to-speech in Google AI Edge Gallery, running offline with paired small language models such as Gemini Nano and Gemma 4. Google says ML Kit's ASR covers 100+ languages at 10x real-time speed, and its TTS begins speaking in about 100 ms. Open-weight speech bundles for MediaPipe Tasks and LiteRT, plus open-source reference code, are also available.

Heat Trend

Not enough continuous observations to show a trend yet.