How We Got Voice AI Response Times Under 1.8 Seconds on Real Phone Calls

<p>Most developers underestimate how hard voice AI actually is.</p> <p>To build a production-ready calling agent, you need to integrate:</p> <p>– SIP signalling</p> <p>– Real-time

<p>Google’s September 15, 2026 announcement of <strong>Gemini 3.8 Live</strong> and <strong>Gemini 3.8 Live Extended Thinking</strong> is easy to skim as another model-version

<p>A voice agent that felt broken on every first turn, and the latency budget I had to take apart stage by stage to find the two dead seconds.</p> <p>The bug report was one

<p>I spent <strong>3-4 weeks</strong> building an AI voice agent that cold-calls HVAC owners for me. Real outbound. Real objection handling. Real appointments booked. When we

<p>When I was building <a href="https://intervu.dev" rel="noopener noreferrer">intervu.dev</a>, an AI mock coding interviewer that conducts full voice interviews in the browser,

<p>Most voice AI teams optimize for one thing: speed. Fewer are asking what happens when the network itself is the problem, not the model.</p> <p>We wrote up an approach to this:

<p>I've been running <a href="https://github.com/openclaw/openclaw" rel="noopener noreferrer">OpenClaw</a> as my personal AI assistant for a while — text-based, the usual way. It

<p>Most AI tutors today feel like glorified chatbots with a text-to-speech layer slapped on top. But what if students could actually <em>talk</em> to their tutor, pause