Unverified50% confidenceReleaseExact time
The Block team released Dessa, which it officially calls "the most realistic full-duplex speech model to date".
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/24/2026
First Seen
Valid until: 11/7/2026
Sources
Related Entities
Related Claims
UnverifiedTreble focuses on the development of a voice simulation platform.54% similarUnverifiedNVIDIA's Nemotron series includes a full-duplex speech model that is a single speech-in, speech-out model, abandoning the traditional ASR→LLM→TTS three-stage pipeline and supporting interruption at any time.53% similarUnverified全双工语音将成为未来所有AI语音产品的标配53% similarUnverifiedDescript has a built-in Overdub feature that, after recording a sufficient voice sample, can synthesize the user's own AI voice to patch recordings by typing.53% similarUnverifiedThis voice streaming model ranks first in both quality and speed (as claimed by the vendor).52% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1092486API
curl https://kongchang.com/api/v1/knowledge/claims/1092486MCP
get_claim(id=1092486)