Browse Comments — Raw (as collected)

Our ML team is incredible! 📈

Congratulations. Your still my favorite in the audio space. Might want to try RuVector by Reuven Cohen I've been trying to get it to work and can't get through the hype of it but if the claims are partially true this could push you further.

Sounds amazing Mike

As somebody that lives & breathes subtext in communication, I'm so excited to dive into this head-first. Thank you & the team!

Ben Bartolone Excited to have you try to beat the demo :)

Go Modulate!

I am confused. So you built a voice model that uses hundreds of other models to understand voice? Your comparison at the beginning with deepseek is what is throwing me off.

Congrats Mike Pappas ! How can I try EML?

Congratulations to the whole team behind this!

Exciting.

Impressive work - development of the Ensemble Listening Model may be a game changer in the AI landscape. It's exciting to see such innovative solutions that not only enhance efficiency but also preserve the richness of human communication.

This is really cool! love seeing AI evolve beyond just scaling Mike Pappas

The point about voice getting flattened into text really lands. f meaning lives in nuance and context, it makes sense that one giant model isn’t always the right tool for the job. Mike Pappas

Ensembles over brute-force scaling feels inevitable. If this holds up in production, it’s a big rethink of the current AI playbook.

David Boyle

It's a model of models! I like it. Very excited to see where this is going. Great graphic by the way.

Me hearing about a new AI architecture

About to deep dive the research, how does this differ from MoE architecture though? Sounds super exciting.

Amazing!

This is the shift from bigger is better to smarter is faster.