Hacker News Reader

sarangzambareon Show HN: Dia, an open-weights TTS model for generating realistic dialogue

no worries, i will email you

sarangzambareon Show HN: Dia, an open-weights TTS model for generating realistic dialogue

Impressive demo! We'd love to use this at https://useponder.ai

time to first audio is something that is crucial for us to reduce the latency - wondering if dia works with output streaming?

the python code snippet seems to imply that the entire audio bytes are generated directly?

sarangzambareon Show HN: SuperCraft – Turn industrial design sketches to 3D renders

Great feedback about the iteration using conversation.

edit 1 : You are spot on about the infinite child nodes - working to add that currently, just wanted a first MVP in people's hands as fast as possible and iterate.

edit 2: that's strange, i just tried sharing a board of my own ->

https://supercraft.ai/public_board?boardId=66c88341614fe5ecb...

i am able to see this in an incognito window, can you see it?