LiveKit: Powering Real-Time Communication - Including OpenAI Voice Mode

Real-Time Data Transmission Solutions

Many technology businesses encounter difficulties when attempting to deliver high-bandwidth, multimodal data – such as concurrent audio and video streams – to users in real time, without experiencing disruptions. While some organizations opt to develop internal solutions, these frequently necessitate substantial ongoing maintenance and support.

The Creation of LiveKit

Recognizing this challenge, Russ d’Sa and David Zhao developed LiveKit, an open-source software suite designed for constructing applications capable of transmitting real-time audio and video. The project was initiated in 2021, and its creators quickly identified potential commercial viability.

Rapid Growth and Adoption

Their assessment proved accurate. According to d’Sa, LiveKit currently boasts “over 500 paying customers and more than 100,000 developers utilizing both its cloud platform and open-source offerings.” The platform serves as the “backbone for approximately 25% of 911 emergency calls within the U.S.”

Furthermore, it is employed by major aerospace corporations for launch and flight monitoring, by Skydio for remote operation of police drones, and by teams at Oracle and Adobe in diverse governmental applications.

From Experimentation to Startup

The impetus for transforming LiveKit into a business arose when “prominent companies including Spotify, Oracle, and Reddit began experimenting with LiveKit and requested a cloud-hosted version,” d’Sa explained to TechCrunch. He described the vision as “Cloudflare, but specifically for media streaming.”

Consequently, d’Sa, a former engineer at Twitter, and Zhao, previously the director of engineering at Motorola, decided to establish LiveKit as a startup and introduce a managed service: LiveKit Cloud.

Current Offerings and Clients

Today, LiveKit, which also underpins OpenAI’s ChatGPT Voice Mode, provides Software Development Kits (SDKs), tools, and Application Programming Interfaces (APIs) that empower developers and companies to create immersive streaming video and audio experiences.

The startup’s clientele includes technology leaders such as Spotify, Meta, and Microsoft, alongside companies like Character AI, Speak, and Fanatics.

Future Development and Expansion

The San Jose, California-based company is presently focused on expanding its engineering and product teams – currently numbering around 50 employees – and enhancing its fundamental infrastructure.

LiveKit is also in the process of developing what d’Sa refers to as an “elastic agent compute service,” a product designed to deploy and automatically scale voice “agents,” such as chatbots, based on demand.

The Vision for an AI-Native Cloud

“Ultimately, LiveKit is evolving into ‘AIWS’ – an AI-native cloud provider,” d’Sa stated. “In the same way that Stripe revolutionized payments, LiveKit aims to transform communications.”

Financial Performance and Funding

LiveKit’s financial standing is strong, according to d’Sa. The company achieved a run rate exceeding $10 million last year.

Recently, LiveKit secured $45 million in a Series B funding round, spearheaded by Altimeter, with contributions from Redpoint Ventures and Hanabi Capital.