Sopro TTS: A 169M model with zero-shot voice cloning that runs on the CPU

Re:Sopro TTS: A 169M model with zero-shot voice cloning that... by onion2k on Friday January 09, 2026 @03:31AM (#46551363)

zero-shot is a single prompt (maybe with additional context in the form of files.)

few-shot is providing a few examples to steer the LLM

multi-shot is a longer cycle of prompts and refinement

[ Parent | on HN ]

Re:Sopro TTS: A 169M model with zero-shot voice cloning that... by moffkalast on Friday January 09, 2026 @05:36AM (#46552320)

if you had one-shot

or one opportunity

[ Parent | on HN ]

Re:Sopro TTS: A 169M model with zero-shot voice cloning that... by nake89 on Friday January 09, 2026 @05:41AM (#46552360)

to seize everything you ever wanted in one moment

[ Parent | on HN ]

Re:Sopro TTS: A 169M model with zero-shot voice cloning that... by raccer on Friday January 09, 2026 @11:13AM (#46555376)

Would you capture it or just let it slip? Yo

[ Parent | on HN ]

Re:Sopro TTS: A 169M model with zero-shot voice cloning that... by mikkupikku on Friday January 09, 2026 @08:39AM (#46553703)

I've been calling good results from a single prompt "single-shot." Is this not right?

[ Parent | on HN ]

Re:Sopro TTS: A 169M model with zero-shot voice cloning that... by flimflamm on Friday January 09, 2026 @12:00PM (#46555968)

No as you could have multiple examples of expected output in single prompt. You should just "ask" - that's zero shot. If you "ask + provide" examples then you are in the n+1 shot realm. But I suppose terminology is shifting from this...

[ Parent | on HN ]

Re:Sopro TTS: A 169M model with zero-shot voice cloning that... by samtheprogram on Friday January 09, 2026 @09:24AM (#46554149)

This is one-shot.

[ Parent | on HN ]