Complete AI Training
Start for Free

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI news ·

Zyphra Releases Open ZONOS3-Mini Voice-Cloning Model

Zyphra released an open voice-cloning model that runs on everyday devices and gives users control over speech traits, while also raising the need for consent and verification safeguards.

1 min read
CreativesEducationMarketingProduct development

Zyphra has released ZONOS3-Mini, an open voice-cloning model now available under the Apache 2.0 license. The model is designed for zero-shot voice cloning and can run directly in browsers, on mobile phones, and on local computers. Its sparse architecture provides controls for speaker characteristics, pacing, and emotional tone while maintaining low reported latency.

The release broadens access to voice-cloning tools for professionals in creative fields, education, marketing, product development, science, and writing. At the same time, the wider availability of the technology raises the need for consent mechanisms and provenance safeguards to verify the origin of synthetic speech.

Share