Google AI Edge Gallery: offline speech recognition and synthesis on-device
Google added fully offline speech recognition and text-to-speech to AI Edge Gallery, with ML Kit APIs on Android and MediaPipe/LiteRT models cross-platform, pairing with on-device small language models.
Why it matters
Voice in and voice out with zero cloud round-trips: 100+ language ASR at 10x real-time and a diffusion TTS that starts speaking in about 100ms, all running on the phone.
What you could build with it
A developer building a language-learning app could add offline voice chat to it: the learner speaks, on-device speech recognition transcribes, a small on-device model answers, and text-to-speech reads the reply aloud, all working on a plane with no connection.
Does it hold up?
Too early to judge: the feature shipped Nov 10 in the Gallery app and ML Kit APIs; no independent app builds or benchmarks have appeared yet.
Built with Google AI Edge Gallery
- Google AI Edge Gallery adds offline on-device speech recognitionarticle · Writeup of the release: ML Kit ASR/TTS on Android, Parakeet-TDT, PocketTTS, Kokoro and Phonon cross-platform, plus the XR Language Teacher demo.
- On-device speech recognition and text-to-speech in Google AI Edge Galleryarticle · Google's own writeup: ML Kit ASR/TTS on Android, Parakeet-TDT/PocketTTS/Kokoro/Phonon cross-platform, plus the XR Language Teacher showcase and streaming voice dictation in AI Chat.
- Google AI Edge Gallery: how Google quietly put a full AI lab in your pocketarticle · Independent overview: from a sideloaded alpha to a Play Store app with 1M+ downloads and 23,000+ GitHub stars, with AI Chat, Audio Scribe and on-device benchmarks.
Learn more
First spotted on github: source.
More AI models
VoiceStudio: open-source, fully-local ElevenLabs alternative for voice cloning, dubbing, and audiobooksdebpalash/VoiceStudio is a desktop app that runs voice cloning, voice design, video dubbing, dictation, transcription…models · JEV 0.75Edge0-145B: open 145B MoE that runs on a single consumer GPUEdge0 open-sourced 145B and 120B latent-MoE models derived from Qwen3.8-Flash-Next, plus an inference framework that…models · JEV 0.74Strata: run a 125B model on a gaming PCFree, MIT-licensed local inference engine that runs Qwen3.8-Flash-Next (125B MoE, about 6B active per token) on…models · JEV 0.73Soup: fine-tune an 8B LLM on a 4 GB laptop GPU from one YAML fileAn open-source CLI that fine-tunes LLMs from a single YAML file, with layer streaming that trains an 8B model on a 4…models · JEV 0.73Tencent Hunyuan Hy3: open 295B MoE with 21B active under Apache 2.0Tencent open-sourced Hy3, a 295B-parameter mixture-of-experts model with only 21B parameters active per token, under…models · JEV 0.73NVIDIA Isaac Lab 3.0 GA + Isaac OS open runtime + Agile ONE SN humanoidMajor Isaac platform expansion at GTC Berlin Oct 21: Isaac Lab 3.0 is generally available as the open-source robot…models · JEV 0.73
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.