Zyphra releases open-source real-time voice cloning model ZONOS2
Original title:ZONOS2: Real-time TTS with High-Fidelity Voice Cloning
AI summary
Zyphra releases ZONOS2, an Apache 2.0-licensed real-time TTS model built on an MoE architecture with 8B total parameters and 900M active parameters, focused on high-fidelity zero-shot voice cloning, with weights now available on Hugging Face.
Key Takeaways
- Zyphra releases ZONOS2,an Apache 2.
- 0-licensed real-time TTS model built on an MoE architecture with 8B total parameters and 900M…
- focused on high-fidelity zero-shot voice cloning,with weights now available on Hugging Face.
Quick Facts
- Published
- Updated
- Source
- Zyphra Research(网页)
- Category
- Model Releases
- AI score
- 71
原文给出架构、训练数据和评测细节,读者可以了解这个开源实时 TTS 模型在音色保真与生成稳定性之间的取舍。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Zyphra releases open-source real-time voice cloning model ZONOS2. June 12, 2026. https://new.nat6.net/en/news/cmtym34cc000aro1n6plzqbfe/