Bland unveils Tofu-1, a full-duplex speech-to-speech model built for noisy high-stakes calls
Scobleizer · x · 2026-11-14
Bland introduced Tofu-1, a full-duplex speech-to-speech model now open for an enterprise research preview. It listens while speaking, lets callers interrupt mid-sentence, separates voices from noise, distinguishes multiple speakers, and can even generate entire audio scenes including overlapping voices and ambient sound—every sound in the demo video was model-generated. It's trained on the messy real-world calls enterprises actually receive: storms, weak signal, crying kids, several people talking at once—and reasons while talking without claiming a task is done prematurely.
Related event: Bland launches full-duplex speech model Tofu-1 for noisy, high-stakes calls(2 posts)→
More from coding & agent
- HQ rebuilds local bot runtime: self-host agents, chat from desktop, Slack or Telegram — jacob_posel · 2026-11-14
- HQ adds MCP signup: create an account inside ChatGPT, Claude or Grok — jacob_posel · 2026-11-14
- Coolify TLS cert rotation works, but CA API buckles under eMudhra fallout rush — tristanbob · 2026-11-14
- Delta editor adds Claude Agent via ACP, works with existing Claude subscriptions — dean_rie · 2026-11-14
- Claude Code 2.1.324 is about to be released — ClaudeCodeLog · 2026-11-14
- Matthew Berman open-sources Scarlet, an AI video editing agent that speeds up his workflow — MatthewBerman · 2026-11-14