LiveKit, the infrastructure platform powering OpenAI’s ChatGPT voice mode, has closed a $100 million Series C funding round at a $1 billion valuation. Index Ventures led the investment, with participation from Salesforce Ventures, Hanabi Capital, and returning investors Altimeter and Redpoint Ventures.

The funding arrives just ten months after LiveKit’s previous round and shows how quickly companies are adopting voice AI technology. The company serves over 200,000 developers and teams globally, including major customers like Tesla, xAI, Salesforce, 911 emergency services, and mental health providers.

Voice AI Moves from Novelty to Production Scale

Voice interaction represents a major change in how we use computers. For the first time, machines can hold conversations the same way people do. What began as experimental features inside products like ChatGPT has expanded into real applications across banking, healthcare, retail, customer support, education, and robotics.

Big companies are moving fast. Salesforce’s voice agents now handle customer support for major global brands. Tesla uses voice AI across its sales, support, insurance, and roadside assistance. Many companies are still testing the technology, but several have already launched it at scale with real customers.

Experts expect 2026 to be the year voice AI becomes widespread across thousands of commercial applications. But making this happen requires completely different technology than what powers traditional websites and apps.

Building the Technology for Real-Time Conversations

Traditional website technology was designed to move text between computers. Each action happens independently without remembering what came before. When apps need to remember something, they fetch it from a database, which adds delays. That’s fine for browsing websites but doesn’t work for natural conversation.

Voice AI applications work differently. They operate in real-time and need to remember everything. Conversations might last minutes or hours, with AI agents continuously listening, thinking, and responding while remembering the entire discussion. This difference changes everything about how the technology works.

You can’t build, test, launch, or monitor voice applications using normal website development tools. The entire technology foundation needs rebuilding to support real-time systems that work like human conversations. LiveKit is building that foundation with all the pieces designed to work together.

How the Series C Funds Will Accelerate Development

The new funding will help LiveKit move faster on its main goal: making voice AI development as easy as building websites. The company is expanding its platform to cover everything developers need, from basic tools and networking to AI model management, phone integration, and performance monitoring.

LiveKit plans to expand its global network, which already handles billions of calls every year between voice agents and users. The company recently partnered with phone carriers worldwide to connect its network directly to telephone systems, making phone-based voice experiences faster. More investment here will improve performance and reach more locations.

The money will also support LiveKit’s strategy of hosting AI models close to where the voice agents run. This reduces delays and complexity that happen when agents need to contact distant cloud services from AI model providers.

From Open Source Project to Billion-Dollar Platform

Russ d’Sa and David Zhao founded LiveKit in 2021 as a free, open-source project for building apps that could transmit audio and video in real-time without interruptions. The timing matched perfectly with the pandemic, when everyone was suddenly meeting on Zoom.

What started as a free tool for developers evolved when the founders saw that big companies wanted professionally managed versions of the technology. The voice AI boom accelerated that change. Their different but complementary skills—d’Sa’s vision for natural human-computer interaction and Zhao’s technical expertise—have driven the platform forward. The two founders have worked together for nearly 20 years across multiple companies.

LiveKit now covers the complete process of building voice agents. The company offers tools for every major platform. LiveKit Agents, built from their work on ChatGPT Voice Mode and downloaded over one million times monthly, gives developers control over how agents work with connections to hundreds of AI models. The system automatically handles conversation details like knowing when to speak and when to listen.

Agent Builder provides a no-code option for teams that want to start with templates, adjust prompts, map out workflows, and share prototypes before writing any code.

Solving the Testing and Monitoring Challenge

AI models don’t give the same answer twice, even with identical questions. This makes traditional software testing methods useless. You can’t check if something works correctly when it behaves differently every time. LiveKit’s solution uses statistical testing methods similar to how we evaluate people through exams.

LiveKit Agents supports proper testing of agent code, with detailed tracking that connects to standard monitoring tools for deeper analysis. For simulation—essentially testing how well things work together—the company partnered with Bluejay, Hamming, and Roark. These simulations run thousands of conversations between voice agents, changing prompts, languages, and voice styles to build confidence the system works reliably.

Agent Observability solves another problem. No tool existed before to understand what’s happening during live calls between agents and users. Developers couldn’t easily answer basic questions: How long did the agent take to answer? What did it understand when users said medication names? Did users press zero to reach a human? How fast was the conversation?

LiveKit’s monitoring platform answers these questions through recorded sessions, detailed tracking, synchronized transcripts, and error reports. With these insights, developers can go back and improve their agents.

Investor Perspective on the Infrastructure Opportunity

The Index Ventures partner leading the round highlighted the technical skill and product sense of LiveKit’s founders. After years working with developer-focused, open-source infrastructure companies, the investor noted how rare it is to see that combination.

The platform solves a fundamental networking and computing challenge. More than 200,000 developers already use LiveKit, with growth happening almost entirely through developers choosing it rather than traditional sales teams. Customer feedback consistently mentions how developer-friendly the platform is, how much control it gives over scaling, and how much time it saves.

As AI models become commodities, the value moves to infrastructure companies that solve complex challenges in simple ways. LiveKit’s complete approach—covering development tools, networking, platform services, and full backend management for real-time AI—positions it to capture major value.

The investment thinking goes beyond current voice agent applications. As robots and autonomous systems advance, they’ll need the same real-time infrastructure to interact with the physical world through cameras, microphones, and sensors.

Origianl Creator: Ekaterina Pisareva
Original Link: https://justainews.com/applications/voice-and-speech-recognition/livekit-hits-1b-valuation-with-100m-series-c-from-index-ventures/
Originally Posted: Fri, 23 Jan 2026 07:02:42 +0000