Corroborated event · 3 outlets
Block introduced Dessa, a speech model trained on human speech that processes audio directly and is designed to recognize emotions, sarcasm and pauses in real time.
AI brief
Block introduced a voice model called Dessa, which it says draws its knowledge from conversational data rather than being paired with a large language model for retrieval.
Written by AI from TechCrunch AI's published text.
More coverage of this story (2)
Block introduces Dessa AI voice model
Block's new AI model Dessa was trained entirely on human speech to catch emotions text-based models miss
The Decoder · 24 Oct 2026, 14:09
Jack Dorsey’s Block made an AI model that talks to you like an awkward new friend
The Verge AI · 23 Oct 2026, 19:32
Primary source
“Typically, speech models are paired with a large language model for knowledge retrieval. Block said that Dessa's knowledge stems from the conversational data it has.”
Reporting from a publisher outside the organisation being covered.
Read original story ↗