I'll get straight to the point.
We trained 2 new models. Like BERT, but modern. ModernBERT.
Not some hypey GenAI thing, but a proper workhorse model, for retrieval, classification, etc. Real practical stuff.
It's much faster, more accurate, longer context, and more useful. 🧵
Thanks to everyone who attended my talk on "Gathering Insights from Audio Data" at #VOICE22 Here's a link to the slides that I presented: https://t.co/ZsRY0IFZgQ
Please reach out if with any questions. Thanks!