Show HN: Caramel: a speech-to-speech conversation model trained only on audio
0 comments
- Points4
- Comments0
0 comments
I trained a speech-to-speech model last night, inspired by Dessa. I'm calling the model caramel-1. The model learned from recordings of people talking and nothing else, and has never seen a single word of text. The code, the training recipe and the weights are open source. caramel-1 sounds incredibly human. Most of what caramel-1 says doesn't make sense yet, but the sentences make a little more sense every time I add more data. Caramel is fast. On a GPU, caramel-1 makes 1 second of audio in…
Story details from Hacker News ↗ · linked article on naklecha.com.
Show HN: Caramel: a speech-to-speech conversation model trained only on audio was first seen on Hacker News · Show HN on Nov 11, 2026. Since then it has appeared in 9 checks on 1 different day. Its best position was #24 of 30, reached on Nov 11, 10:21 UTC. It dropped off the list after Nov 11, 14:21 UTC.
Position on Hacker News · Show HN at every check (lower is better).
Best rank per day (UTC).
Platform metric as reported by Hacker News.
Archived daily, weekly and monthly lists that include this entry.
Not at the moment. It was last on the list on Nov 11, 14:21 UTC, after 1 day of trending.
#24 on Hacker News · Show HN, on Nov 11, 10:21 UTC. Trendoday checks this list every 30 minutes.
It was first seen on Nov 11, 2026 and has been on the list on 1 day (9 checks) since.
4 points and 0 comments; submitted by naklecha on Nov 11, 2026.