🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Soon it will seem so strange that we had to click around websites to find what we were looking for because we'll be used to just asking for what we want. Young people will scrunch their nose at you when they see you using a mouse and keyboard!

Finally!

Excited for the next LiveKit event! I always make new friends at these and learn a ton!

Faster, more accurate, cheaper. The new AssemblyAI streaming model is an exciting advancement in voice AI agents! It’s easy to try it out in LiveKit! assemblyai.com/docs/speech-to…

Interruptions are the biggest problem with Voice AI agents. I'm giving a talk on the academic research on turn taking in human conversation, and the SOTA end of turn models used in Voice AI, at the AI Engineer World's Fair tomorrow. Hope to see you there!

Magic.

At #CCW and interested in learning more about Voice AI agents? Join LiveKit and AssemblyAI for a happy hour at CCW tomorrow (Tuesday) June 10th from 4PM-6PM to talk about all things Voice AI! Very limited space so RSVP now below to secure your spot! lu.ma/od8f45kf

If you’re at #ccw and want to learn about voice ai from the people who helped build OpenAI’s voice mode, DM me or find me by the registration escalators. I’d love to meet you!

I feel personally emboldened by all the people at this professional conference wearing shorts.

Since joining LiveKit, a lot of Voice AI devs have asked me why they should use WebRTC instead of WebSockets. The question comes up a lot, especially in the context of telephony agents that use Twilio (Twilio streams the audio using WebSockets). Here's why.

Ink-Whisper is fast! Cartesia released a new STT model yesterday, and it's as fast as you'd expect. Streamed transcription finishes in <100ms—before you can say the next sentence. I used it to build a live teleprompter that follows along with what you're saying. Code in 🧵

I'm personally in the dark mode camp, but a lot of users were asking for this (I don't understand why). It's fun being on a team that ships at such a fast rate!

Join us on June 17 for LiveKit's Realtime AI Meetup! Our events are a great way to meet other people building in Voice AI! lu.ma/xzpxc6uf

🚀 Deploying a real-time AI voice agent with ~500ms latency — globally. We just dropped a full breakdown of how to assemble a low-latency speech pipeline using STT, LLMs, and TTS, optimized for production use. Here’s how we did it with LiveKit 🧵

We just made state-of-the-art TTS 20x more affordable. $5 per million characters. And we're open sourcing the training and modeling code (built on Llama). Because scaling voice AI shouldn't break your budget. Technical Details → Inworld.ai/blog/introduci… Why and how we did it

I tried on my kid's mood ring and it got stuck. Here were the sequence of colors it displayed: 1. Green (calm) 2. Amber (nervous) 3. Grey (strained) 4. Black (panicked)

I loved Thanh-y Nguyen's video on building your own Jarvis with LiveKit! It thoroughly shows what anyone can build with modern voice agents. No wonder it got 33k+ views and pushed our internal count of LiveKit Cloud projects named 'Jarvis' over 1,000! youtube.com/watch?v=An4NwL…

Animals responsible for killing the most people every year: 1. Mosquitoes 2. Other humans 3. Vending machines

#AmazonNova Sonic now integrates with LiveKit’s WebRTC infrastructure 🚀🎉⚡ This integration streamlines the development of voice-first AI application by removing the complexity of managing real-time audio infrastructure. #AWS #generativeAI 👉 go.aws/4lhtJ4P

This is so much fun!