LaunchAI ModelsNovember 15, 2026

Bland launches Tofu-1 full-duplex speech-to-speech model for calls

Read original source →marktechpost.com

Tofu-1 is in research preview and handles audio in and out with no transcript in between, targeting calls from moving cars, hospital rooms and weak-signal lines. Bland lists speaker tracking, noise separation and paralinguistics as its three audio skills; internal architecture is unpublished.

2 sources

More stories today

Open the live feed