🎙️ Will conversational AI take over our devices? Patrick Ferriter from @AgoraIO joins the IoT For All #podcast to discuss building real-time #AI voice experiences, even on bad networks. ➡️ youtu.be/qEBnlgaEUBY

We’re excited to announce that @AgoraIO has selected Soniox as a speech-to-text provider for its real-time Conversational AI platform. For over a decade, Agora has built the global network infrastructure that connects people through real-time voice, video, and messaging. Today, that same carrier-grade infrastructure is powering a new era of interactions: humans talking directly with AI. For voice AI, moving audio in real time is only part of the problem. The system must understand natural, conversational speech instantly—accurately capturing critical details despite accents and noisy environments. Soniox brings: • Native-speaker accuracy across 60+ languages • Ultra-low latency for real-time conversations • Seamless multilingual speech and mid-sentence language switching • High precision on names, numbers, IDs, and other critical details • Reliable performance on real-world audio at production scale Agora provides the orchestration for Conversational AI and the infrastructure for real-time communications. Soniox provides the intelligence for understanding spoken language. Together, we’re making it easier for developers to build voice AI that can communicate naturally with people around the world. Excited to partner with the Agora team and help power the next generation of conversational AI.

One really cool use case I’ve been exploring with Jev: a semantic VAD for real-time voice AI. I plugged it into @AgoraIO ConvoAI and used the live transcript + recent conversation as context to decide whether the user has actually finished speaking. That makes Jev surprisingly good at handling things like hesitation, unfinished thoughts, and those moments where you stop talking for a second because you’re still figuring out what you want to say. It’s a nice example of where Jev fits really well in a real-time AI loop. Here’s what it looks like in action 👇

Create and deploy custom audio with our new text-to-speech models: 🔵 Gemini 3.8 Flash TTS: Design unique voices with distinct accents and characteristics. 🔵 Gemini 3.8 Flash-Lite TTS: Built for efficiency and scale, choose from your created styles or our expansive production-ready library.

🗣️ Introducing Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking. These advanced audio models are built for natural conversation, featuring major upgrades in turn-taking and near real-time reasoning. They also significantly streamline how you build intelligent voice agents. Until now, advanced reasoning and reliability came at the cost of high latency and the complexity of cascaded, multi-model pipelines. Stitching together separate speech-to-text, reasoning, and text-to-speech models adds delay and inflates costs. By replacing that stack with a single multimodal API call, these two models make building highly responsive voice agents simpler and cost-effective. 🧵 Here’s what you can build with them: