webmp3/Sakura-sanoTTS-ONNX
Text-to-Speech • Updated • 29
Sakura Audio
Sakura Audio — compact local speech, ASR and TTS models optimized for practical CPU/edge inference.
Note High-fidelity CPU-optimized ONNX & INT8 release of sanoTTS (15 voices, 9 languages)
webmp3/Sakura-Whistle-ONNX
Automatic Speech Recognition • Updated • 48 • 1Note Full CPU-optimized ONNX & INT8 release of Cactus Whistle ASR (125 MB INT8, 54+ TPS CPU)