Cartesia Sonic-3.6与Ink-2刷新语音AI延迟纪录
There is a subtle architecture shift happening in voice AI.
语音交互的核心痛点是延迟，Cartesia这次把听说两端都压到了百毫秒级，做实时语音Agent的同学值得重点关注其架构思路。
There is a subtle architecture shift happening in voice AI.
语音 AI 领域正在发生一场微妙的架构转变。
The voice stack is becoming part of the agent's execution loop.
语音技术栈正逐渐成为智能体执行循环的一部分。
@cartesia is combining the listening and speaking paths around that loop. Sonic-3.6 turns text into speech (90ms latency) and Ink-2 turns speech into text (100ms transcript latency), faster than anything else streaming.
@cartesia 正在围绕该循环整合听与说的路径。 Sonic-3.6 将文本转换为语音（90ms 延迟），Ink-2 将语音转换为文本（100ms 转录延迟），其速度优于所有其他流式传输方案。
now holds the #1 spot for both speaking and listening models.
目前同时占据说话和听力模型的第一名。
Becasue, voice agents are one of those systems where 100ms in the wrong place is very noticeable.
因为，语音智能体是那种在错误位置多出 100ms 就会非常明显的系统。
Cartesia is attacking both sides of that loop at once.
Cartesia 正在同时攻克该循环的两端。
super interesting work by @krandiash and team.
@krandiash 及其团队的工作非常有趣。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力