Maya 2 Native
Maya 2 Native is a streaming text-to-speech model that generates 24 kHz speech in 12 languages, with 20 built-in voices. It includes open weights and inference code, supports zero-shot cross-lingual voice cloning, and is released under Apache-2.0.
What it does
Use it to generate or stream multilingual speech for voice agents, spoken content, and interfaces that can be hosted locally.
- Availability
- Checkpoint release 1.0 · October 25, 2026
- Pricing model
- Not established
- Deployment
- Can be hosted yourself; Linux and an NVIDIA GPU are required for the documented setup
- Privacy
- Not established
- Licence
- Apache-2.0; see component licenses
Capability
Modality
Access
Deployment
Pricing
Target user
Evidence
5 claims- Generates 24 kHz speech in 12 languages with 20 built-in voices. [1]
Built by **Maya Research**, it generates **24 kHz speech** in 12 languages, with **20 built-in voices** and **zero-shot voice cloning from 3–10 seconds of reference audio**.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - Supports zero-shot voice cloning from 3–10 seconds of reference audio. [1]
Built by **Maya Research**, it generates **24 kHz speech** in 12 languages, with **20 built-in voices** and **zero-shot voice cloning from 3–10 seconds of reference audio**.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - Provides Python, an OpenAI-compatible speech endpoint, and WebSocket streaming. [1]
The release includes **open weights and inference code**, with a Python API, an OpenAI-compatible speech endpoint, and WebSocket streaming.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - The release includes open weights and inference code. [1]
The release includes **open weights and inference code**, with a Python API, an OpenAI-compatible speech endpoint, and WebSocket streaming.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - The model is licensed under Apache-2.0. [1]
> **For developers:** the model is **Maya 2 Native**.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights.
Sources
1
1
Hugging FaceOwner-controlled
https://huggingface.co/maya-research/maya2-native
Retrieved 2026-10-28 00:20 UTC
Verified
Versions
1| Version | Effective | Change | Changed fields | State |
|---|---|---|---|---|
| 1 | 2026-10-28 00:21 UTC | publication | None | Current |
In briefs
1- New models bring live speech translation and voice features Daily brief 2026-10-28
Original catalog content is licensed under CC BY 4.0, attributed to wowsuchai. Source quotations and linked content keep their original rights.