Inflect-v2: 3.9M and 9.3M parameter open-weight TTS models

Introducing Inflect-v2, two exceptionally small, open-weight English TTS models at just 3.9M and 9.3M parameters. Both generate speech multiple times faster than real-time on CPU. Despite their size, Inflect-v2 delivers quality that is competitive with much larger lightweight TTS systems, including KittenTTS, Piper, and Supertonic-3.

CPU, CUDA, PyTorch, and ONNX are supported. Apache 2.0.

See it for yourselves: https://huggingface.co/owensong/Inflect-Micro-v2 https://huggingface.co/owensong/Inflect-Nano-v2

Try the Demos: https://huggingface.co/spaces/Nymbo/Inflect-TTS (unlimited CPU usage) https://huggingface.co/spaces/owensong/Inflect-v2 (ultra-fast ZeroGPU usage)