Product LaunchTechCrunch

Block’s new voice model sounds eerily human

#voice ai#block#speech model#dessa#jack dorsey

English

Block has introduced a new voice model named Dessa, which operates locally and mimics human speech with emotional expressions such as laughter and pauses. Unlike traditional speech models, Dessa is a full-duplex model that can converse in real-time, trained entirely on human speech without relying on text-based inputs.

中文

Block推出了一种名为Dessa的新语音模型，该模型在本地运行，并模仿人类的语音，具有情感表达，如笑声和停顿。与传统语音模型不同，Dessa是一个全双工模型，可以实时对话，完全基于人类语音训练，而不依赖文本输入。