Unverified50% confidenceFactExact time
The speech model DESA released by a Twitter-affiliated team uses no text pretraining at all, learning to speak by listening to real human conversations, features full-duplex interruption, breathing and laughter, and can run locally on a single Apple computer.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/25/2026
First Seen
Valid until: 1/23/2027
Sources
Related Entities
Related Claims
UnverifiedBlock released DESA, a speech model that supports local operation, can listen and speak in real time and can interject and respond, reportedly trained from scratch using only human speech without relying on text training or pretrained models.74% similarVerifiedApple provides the Speech Framework to support on-device speech recognition, but its capabilities are mainly limited to transcription and lack an LLM polishing step.68% similarUnverifiedIn a pure voice interaction flow there are no visible pop-ups or clickable confirmation buttons, and voice agents often invoke tools and execute actions while the conversation is still ongoing.64% similarUnverifiedTTS zero-shot voice cloning does not train a new model; instead, it encodes the reference voice into a speaker embedding, which is injected as a condition during synthesis, so the same set of weights can provide different timbres for different characters.64% similarUnverifiedVoiceCap promises that recording data will not be used to train models.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1095796API
curl https://kongchang.com/api/v1/knowledge/claims/1095796MCP
get_claim(id=1095796)