Block Unveils Dessa, AI Voice Model Trained Exclusively on Human Speech
Jack Dorsey's company Block has launched Dessa, an AI voice model trained solely on human speech without using text or pretrained models. Dessa can listen and speak simultaneously, recognizing emotions, sarcasm, and natural pauses, enabling real-time, natural conversations. It processes audio directly, preserving tone and emotional cues often lost in text conversion. The model runs locally on devices like the MacBook Pro M4 Max without internet and is currently in research preview with limited public access.
First-hand measurement across 4 sources
We measured how 4 outlets covered this story. No outlet gave this story a measurable political slant — there is no left–right reading to report. Overall sentiment is positive (72/100). Lens Score 35/100.
Outlets measured: economictimes, indiatoday, thetribune, indianexpress. See how each one headlined and framed the same story in the source comparison below.
AI Analysis
Sentiment was consistent across outlets (65–75/100), indicating broadly factual reporting rather than editorialising.
Coverage timeline
indianexpress broke this story on 24 Oct, 05:45 am. Other outlets followed.