The short version
Compact local text-to-speech model; a user turns PDFs and EPUBs into audiobooks.
From the HackerLinks archive
Kokoro
Compact local text-to-speech model; a user turns PDFs and EPUBs into audiobooks.
At a glance:
First seen:2026-07-07
Last seen:2026-10-21
Times seen:2
Website:huggingface.co
Why it caught our attention
A strong first-hand sound-quality endorsement supports resurfacing after the 90-day cooldown.
Where it surfaced on Hacker News
2026-10-21
“Best sounding local TTS engine I've heard by far is Kokoro It's small enough to be run on modern phones realtime, I use it for listening to PDF's/EPUB as Audiobooks”
gavinray · recommendation · first hand use
Direct HN comment
Voxtral 3
2026-07-07
Editorial paraphrase
HN users said Kokoro is very nice, called it legitimately super good for a tiny model, and shared side projects built on top of it.
Original threadLocal, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro