LIVE
ON-DEVICE VOICE AI

Google AI Edge releases ultra-lightweight offline ASR and diffusion Text-to-Speech runtimes for local mobile Small Language Models

Mobile app developers can now bundle sub-100ms offline speech recognition and expressive diffusion voice synthesis that run 100% locally on Android and iOS NPUs.

UpdateNext Staff
Editorial team
5 min read
Smartphone audio waveform and Google AI Edge on-device neural network processing
Smartphone audio waveform and Google AI Edge on-device neural network processing. Photo via Google Developers Blog / UpdateNext

The missing link for completely offline mobile AI agents has always been the voice layer. While mobile NPUs have grown fast enough to run 3B-to-8B text models locally, expressive voice synthesis typically required streaming audio back and forth to cloud GPUs. This week, Google’s AI Edge team solved that bottleneck.

In a release on the Google Developers Blog, the AI Edge group unveiled production-ready mobile libraries for streaming offline Automatic Speech Recognition (ASR) and lightweight diffusion Text-to-Speech (TTS) across Android (via LiteRT) and iOS.

ADVERTISEMENT
Get the daily AI briefingSubscribe

Full-Duplex Voice AI in Airplane Mode

✓ WE KNOW
  • Google’s AI Edge engineering team has released new open developer runtimes for offline Automatic Speech Recognition (ASR) and diffusion-based Text-to-Speech (TTS).
  • Designed to pair directly with on-device Small Language Models (such as Gemma 3 and quantized Llama/Qwen variants), the pipeline enables full duplex voice conversations in airplane mode.
  • Using 4-bit weight quantization and hardware NPU delegation, the entire speech-to-LLM-to-speech loop runs with under 180 milliseconds of latency.
? STILL UNCLEAR
    “When speech recognition, LLM reasoning, and diffusion voice synthesis all run on the phone’s NPU, voice AI becomes free, instant, and 100% private.”
    —
    SOURCES:Google Developers Blog: AI Edge Offline ASR & Diffusion TTS Release ↗
    > how_we_reported_this
    • ✓Sources: written from primary sources and official statements, linked above.
    • ✓Fact-check: names, dates and claims checked against the original sources.
    • ✓AI use: researched with AI tools; written and reviewed by our editors. Our AI policy
    • ✓Independence: UpdateNext has no financial relationship with the companies covered.
    > react:
    UpdateNext Staff
    The UpdateNext newsroom covers AI and tech from primary sources, every day. Every story is checked against official statements.
    Write for us →
    Comments
    Comments are moderated · Sign in to comment
    > read_next
    2NM MOBILE SILICON
    Qualcomm’s 2nm Snapdragon 8 Elite Extreme Gen 6 runs 30B-parameter MoE AI models locally as Motorola Signature 27 debuts
    GALAXY S27 SHAKEUP
    Samsung Galaxy S27 series schematics leak: New 'Galaxy S27 Pro' brings 200MP camera and privacy display as RAM costs surge
    APPLE IPHONE ROADMAP
    Apple sets spring 2027 launch for 2nm A20 iPhone 18 and iPhone Air 2 while partnering with OpenEvidence on Health AI study
    What's next in AI.

    The daily briefing on AI news, tools and breakthroughs.

    Google AI Edge releases ultra-lightweight offline ASR and diffusion Text-to-Speech runtimes for local mobile Small Language Models | UpdateNext