Neuphonic Releases NeuTTS Air for On-Device AI Speech
The new Apache 2.0 text-to-speech model is built on a Qwen2 architecture and optimized for local inference with GGUF support.
A new open-source text-to-speech (TTS) model called NeuTTS Air has been released by developer Neuphonic. Designed for on-device applications, the model features voice cloning capabilities, allowing users to generate speech in a target voice from just a short audio sample.
Uniquely, NeuTTS Air is built upon a Qwen2 language model architecture, a different approach from many specialized TTS systems. The release includes GGUF-quantized versions, making it compatible with popular local inference frameworks like llama.cpp and accessible on a wide range of consumer hardware without requiring a powerful GPU.
The combination of local execution, voice cloning, and a permissive Apache 2.0 license makes NeuTTS Air a significant new tool for developers. It opens up possibilities for building private-by-design applications, from custom voice assistants to dynamic content creation, without relying on cloud-based APIs and their associated costs and privacy concerns.
Developers can explore the model, listen to audio samples, and download the weights from the project's Hugging Face repository.
Sources
- Visit
neuphonic/neutts-air
Hugging Face
More in Text → Speech
Audio8 debuts a 0.6B multilingual zero-shot TTS preview
The compact text-to-speech model promises voice cloning across languages from a footprint small enough to run without heavy hardware.
KRAFTON releases A.X-K2 Raon speech MoE model
The game maker's new open model blends text-to-speech and speech recognition in a single 21B mixture-of-experts system with just 3B active parameters.
NVIDIA's Audex Unifies Audio Understanding and Speech
A new 30B mixture-of-experts model from NVIDIA handles both listening and speaking within a single audio-text architecture.
0 comments
No comments yet. Be the first to weigh in.