ON-DEVICE SPEECH

Zyphra opens ZONOS3-Mini with compact multilingual speech generation

By IDAKIM Digital · Published November 13, 2026

The release combines steerable text-to-speech, voice cloning and browser or native execution options.

Zyphra’s ZONOS3-Mini announcement carries a November 11 original byline and was published on its Hugging Face organization blog on November 12. The text-to-speech model uses 100 million active parameters out of 800 million total and supports 43 languages. Apache 2.0 model weights accompany MIT-licensed runtime and reference code, with paths for WebGPU browsers, Apple Silicon and standalone native inference. The company describes zero-shot voice cloning and control over delivery, plus hardware-specific latency demonstrations. Those figures do not establish the same responsiveness on every phone or browser. The compact active model is relevant to local voice interfaces and private narration workflows, but quality still depends on language, reference audio and the chosen runtime. A source release also does not confer rights to imitate someone’s voice.

Why it matters

Preview a short script using an authorized voice. Compare pronunciation, prosody and first-audio latency on the actual target device before rendering a large batch.

Source: Zyphra · November 11–12, 2026

Product claims are attributed to the linked publisher. IDAKIM Digital has not independently tested these claims.