Block unveils Dessa, a speech-only AI model that reads emotion

Block's Dessa arrives after several attempts at emotionally expressive voice AI that haven't stuck. Hume AI launched EVI in March 2024, Kyutai followed with Moshi in September 2024, LAION's Open Empathic project has been working since late 2023 on empathetic speech synthesis using an open-source audio dataset, and OpenAI overhauled ChatGPT's Advanced Voice Mode in June 2025 with better intonation, pausing, and more realistic expressions of empathy and sarcasm. Block's distinction is architectural: Dessa is trained entirely on human speech with no text data at all, built from scratch, so it processes audio directly rather than converting it to text first. That lets it pick up on emotions, sarcasm, laughter, and pauses in real time, and it supports full-duplex communication so it can listen and talk at the same time. Block says it built Dessa out of curiosity, a desire for better human-AI communication, and safety concerns, since current voice AI ignores emotional cues. Sounding emotional isn't the same as being emotional or intelligent, and for advanced speech AI, capabilities like reasoning while responding matter more than how natural the voice sounds. Dessa isn't publicly available yet; Block is first looking for AI security researchers to test the model for unexpected risks, with sign-up via a form.