skinny.

Latest / Kana & Mari's SoundRepos (English)

sahilmahendrakar/paradee | small English text to speech model | GitHub

Paradee is a compact English text-to-speech model distilled from Kokoro-82M, with four American and British voices. It generates 24 kHz audio using a small ONNX model and supports Python, a command-line interface, and browser use. Related: sahilmahendrakar/paradee / small English text to speech model / GitHub

The skinny

The skinny isn't ready yet — notes appear once the transcript is processed.