Hacker News Reader

Sounds natural and intelligible at 3x speed, which is plus.

>The Real-Time Factor (RTF) is around 0.65.

What is the state of real time tts models?

0.65 isn't that big a gap. This is probably less than one jart away from being optimised to real-time.

Aura can stream responses and has a time-to-first-byte around 200-400ms