Kale TTS runs 65-180x realtime on laptop CPU, deploys to Raspberry Pi and phones

NERDDISCO · x · 2026-11-20

Kale TTS models run anywhere onnxruntime does — Python, Node, browser (WASM and WebGPU), C, Go, Rust, Swift — hitting 65-180x realtime on a laptop CPU. With one extra export file they also load into sherpa-onnx and kokoro-onnx, covering Android, iOS, Raspberry Pi and Home Assistant.

Original post →

More from Infra

Infra channel →