LiveKit
@livekit

Open source framework and cloud platform for building voice, video, and physical AI agents. livekit.com

 🌐
github.com/livekit/agents
Joined June 2021
Meet SpookyKit, five haunted voices for your voice agent 🎃 Voices from LiveKit coworkers (guess who?), via @FishAudio as a custom voice: 🕯️ the hollow whisperer (v_rcMdmMksGhqr) 📞 the grave caller (v_ou7cbav4LMuq) 🌊 the thing below (v_U9DZ7HDMqqtM) Voices summoned from the shadows by our spooky friends at @GradiumAI 🧡 💀 the bone barker (hwchrc8s07s20000) 📻 the dead air host (hwchrc8s07s00000) paste a voice ID into LiveKit Inference or Agent Builder and go give someone a fright 👻 Try them now > cloud.livekit.io
6
4
50
7,484
Voice agents get built three ways: cascaded (STT, LLM, TTS), turn-based speech-to-speech, and full-duplex, where the model listens and talks at the same time. All three now run on LiveKit Inference: GPT-Live and GPT Realtime from @OpenAI, @grok Voice Think Fast from @xai, and Gemini Live from @GoogleDeepMind. One LiveKit Cloud key, one bill.
16
16
200
18,872
😁
Ian Chan
@chanian
Oct 13
Replying to @IanLandsman
1
8
949
Next week: AI Agents Speakeasy w/ @modal, #LATechWeek. Cocktails, delicious food, and conversation with the people shipping agents. Wed Oct 14 · 6-9pm · LA RSVP: partiful.com/e/4K1RcwGixzueN…
4
13
1,094
Know exactly what your agent can do, catch gaps before customers do, and test any model against your own scenarios before you switch. Your agent has a QA team now. Try LiveKit Simulations free through October: livekit.com/products/agent-s…
5
4
42
2,437
Listen to this support call using @MicrosoftAI's speech models and LiveKit Agents. • MAI-Transcribe-2-Streaming hears you • Gemma 4 on LiveKit Inference reasons and calls tools • MAI-Voice-2.1-Flash answers What do you think? STT: livek.it/E4qoZvU TTS: livek.it/RsCc37l
5
5
78
5,597
@AssemblyAI Universal 3.6 Pro is live in LiveKit Inference! - 45% fewer wrong yes/no confirmations - ~30% less background speech - 32 languages + code-switching - Endpointing that waits out phone numbers & emails Swap to universal-3-6-pro, same $0.45/hr Try it now > docs.livekit.io/agents/model…
2
19
1,654
The best robotics conversations don't happen in a keynote hall. Next week we're getting builders in a room instead. Robotics Happy Hour w/ @dimensionalos, #SFTechWeek Tue Oct 6 · 5:30pm · SF RSVP → partiful.com/e/u610QKw40q4lh…
1
1
6
913
Next week: talks and demos on the hard parts of voice AI, then drinks with the engineers actually building it. Voice as an Interface meetup w/ @aquavoice + @speechmatics, #SFTechWeek Wed Oct 7 · 5:30-8:30pm · SF RSVP: partiful.com/e/Qfb44oJOo4cYv…
2
2
7
1,420
Latency. Interruptions. Turn-taking. The hard parts of voice AI, discussed by the people actually solving them. w/ @aquavoice + @speechmatics, #SFTechWeek Wed Oct 7 · 5:30-8:30pm · SF partiful.com/e/Qfb44oJOo4cYv…
4
1
23
1,777
Some of the best conversations at Tech Week happen away from the stage, over a drink, no deck required. AI Agents Speakeasy w/ @modal, #LATechWeek Wed Oct 14 · 6-9pm · LA RSVP → partiful.com/e/4K1RcwGixzueN…
2
11
1,219
Come hear about the latest developments in Robotics and Physical AI from fellow developers, researchers, & hobbyists. LiveKit x @dimensionalos Robotics Happy Hour, #SFTechWeek Tue Oct 6 · 5:30pm · SF RSVP : partiful.com/e/u610QKw40q4lh…
3
1
15
1,196
Enterprise agents need secure access to internal systems. Private links on LiveKit Cloud open a fully-manage, encrypted tunnel straight into your VPC without needing public IPs, inbound firewall rules, or VPN. Live today in the US and EU. Learn more > livekit.com/blog/introducing…
3
4
41
22,190
1 hour of talks + demos on solving voice as an interface. 2 hours of food, drinks, and meeting the engineers who ship this daily. w/ @aquavoice + @speechmatics, #SFTechWeek Wed Oct 7 · 5:30-8:30pm · SF partiful.com/e/Qfb44oJOo4cYv…
2
2
19
1,660
The best ideas still come out of a good conversation over drinks. Join us for the AI Agents Speakeasy w/ @modal, #LATechWeek Wed Oct 14 · 6-9pm · LA RSVP here: partiful.com/e/4K1RcwGixzueN…
1
1
16
1,324
Voice is the interface everyone's racing to get right. Come talk to the people actually building it. LiveKit x @aquavoice x @speechmatics, #SFTechWeek Wed Oct 7 · 5:30-8:30pm · SF RSVP: partiful.com/e/Qfb44oJOo4cYv…
2
3
24
1,909
Your whole voice stack, free for a year on LiveKit. $10K in Cloud credits + a year of Scale + $7K in partner inference credits ($1K each: @googlegemma, @DeepgramAI, @inworld , @rimelabs , @FishAudio, @AssemblyAI, @GradiumAI.) One workspace. No provider keys. Apply now: livekit.com/startups
12
13
126
10,766
Build voice agents that can respond quickly or think longer without changing your stack. Gemini 3.8 Live is available today in LiveKit Agents. Choose gemini-3.8-live for low-latency audio or gemini-3.8-live-extended-thinking for async reasoning. Check the docs for more info: bit.ly/4xWpa6B
15
9
121
51,153
Building a voice agent is a loop: write, deploy, listen to calls, fix what broke. The agent is the unit of work, so we rebuilt LiveKit Cloud's around it. Talk to your agent while you're editing it. Click a metric, see the sessions behind it, go back without retracing. Try it now > cloud.livekit.io/
5
3
68
4,507
We power real-time infra for robots in warehouses, in the air, in your home. We want to meet the people building them in person! Robotics Happy Hour w/ @dimensionalos, #SFTechWeek Tue Oct 6 · 5:30pm · SF RSVP here→ partiful.com/e/u610QKw40q4lh…
4
3
16
1,303
GPT-Live-1 @openaidevs is now supported in LiveKit Agents. It's full-duplex, so it listens and speaks at the same time instead of taking turns, and it hands reasoning and tool calls off to a backend model so a slow lookup doesn't leave your agent sitting in dead air. Same `AgentSession` API. New plugin: `GPTLiveModel`. Get started: docs.livekit.io/agents/model…
25
24
343
85,155
Gradium TTS is now available on LiveKit Inference, free until Oct. 9. Gradium TTS is designed for conversational use cases, with voice cloning, low latency (below 250ms TTFA), and robust pronunciation on difficult cases. Select @GradiumAI as your TTS provider in LiveKit Inference, no additional account or API key required: docs.livekit.io/agents/model…
7
7
61
7,301
Agent deployments are live on LiveKit Cloud. Your voice agent can have a staging deployment running next to production. Same image, same project, same secrets, different name. Test a change on real infrastructure, then promote that exact build. Read more: bit.ly/3V6s89M
4
3
48
4,075
LiveKit Connectors are generally available today. They bring calls from @twilio Programmable Voice and @WhatsApp Business Calling into a LiveKit room as a regular participant, so your agent can answer there with no SIP trunk to provision and nothing new to host. Read the full article: livekit.com/blog/introducing…
6
12
117
8,798
Protect sensitive data without giving up voice agent observability. LiveKit PII Redaction automatically removes personal information from transcripts and recordings before it is stored. It is included with LiveKit Agent Observability at no additional cost. Read more: bit.ly/4gJ1ex7
6
4
57
108,272
1/ Today we're launching Expressive mode in LiveKit Agents. Most voice agents deliver every sentence in that same, upbeat register we all know and hate (including bad news). Expressive mode makes your agent sound like it registers what the caller is actually feeling.
15
17
332
311,301
Here's our take on benchmarks for voice: they're hard to get right, but critical if you're building humanlike AI. Our live benchmarks page is up now: pass rate on real agent scenarios, latency including the worst-case (p99) moments, side-by-side conversation comparison between models over time. Feedback welcome! livekit.com/benchmarks
6
4
63
6,593
The fastest LLM you can put in a voice agent right now is an open-weight model. Gemma 4 31B on LiveKit Inference is only 192ms to first token, 354ms to a full spoken sentence, and cheap. $0.40 per 1M in ($0.20 cached) and $1.20 per 1M out.
7
14
129
9,175
LiveKit Unity SDK 2.0.0 is out of Developer Preview. Bring real-time voice, video, and AI into your Unity game. Build voice chat, live streaming, robot teleoperation, and voice AI agents as NPCs that listen, respond, and react. Give it a try today and let us know what you think.
2
5
54
5,368
Async tools are a new way to handle long-running work in a LiveKit voice agent. Instead of going silent while a slow tool runs, the agent keeps talking: it acknowledges the request, streams progress as each phase finishes, lets the caller cancel mid-task, and won't fire the same action twice. How are you handling tool calls in your voice agents today?
9
5
79
7,412
Frontier models keep getting smarter…and slower. This is bad for voice agents, where every millisecond before the first word counts. So we optimized for speed. Introducing Gemma 4 31B on LiveKit Inference: 🗣️ 381ms to first sentence — 2x+ faster than the next model ✅ 88% task completion (see the criteria in reply) 💰 $1.20 / 1M output tokens Fast enough for real conversation, smart enough for effective agents.
10
19
115
18,170
Join us at the official AI Engineer World's Fair Hackathon in San Francisco, June 27 to 28. 48 hours to bring an idea, build a voice agent on LiveKit, and ship it. Our team will be there to help. Come build something cool. Register here: cerebralvalley.ai/e/aiewf-ha…
3
1
14
1,810
Most voice AI agents forget you the second the call ends. We built a starter kit that fixes it. LiveKit for the voice, @supabase for memory, RAG, and per-user auth, all in one Postgres database. TypeScript and Python included. Give it a try and tell us what you're building.
5
8
82
9,582
Voice AI has a benchmarking problem. Everyone claims their end-of-turn model is the best, but you couldn't actually compare them. Datasets are proprietary, methods are opaque, and there is no shared ground truth. That changes today. We hit this while developing Turn Detector v1, so we open-sourced eot-bench. 5,000+ real user conversation turns across 14 languages, an evaluation harness that measures the real production tradeoff between latency and false cutoffs, and a live public leaderboard. This should become the default way we evaluate turn detection models.
20
11
110
5,787
We shipped LiveKit Turn Detector v1. Instead of reading transcripts, it listens to speech directly, combining semantic and acoustic cues into one end-of-turn prediction. The result: high accuracy, low latency—the best model we tested across 14 languages. Available on LiveKit Cloud.
16
19
162
12,541
Operating a robot over the internet means camera frames and joint state arrive at different times, so your observations drift and training data gets misaligned. LiveKit Portal fuses them back together with the same code, whether the robot's in the next room or another continent.
1
5
42
5,203
We built a live multilingual, multi-person video call with Gemini 3.5 Live Translate on LiveKit. Everyone picks their language, speaks naturally, and hears each other in real time in their language of choice. Watch the demo and check out the open source repo: github.com/livekit-examples/…
Google AI Developers
@googleaidevs
Jun 9
Our latest audio model, Gemini 3.5 Live Translate, takes real-time speech translation to the next level for developers by delivering low-latency translation across 70+ languages. By processing speech as it streams in near real time, the model enables devs to build low-latency audio experiences with: — Multilingual input: Understands multiple languages in a single session without needing to adjust settings. — Auto-detection: Identifies the spoken language and begins translation instantly. — Native audio processing: Generates more natural-sounding speech that preserves speakers' intonation, pacing, and pitch. — Noise robustness: Filters out ambient noise for clearer conversation in loud environments.
11
11
158
21,915
Most voice AI agents forget you the second you hang up. No name, no history, no idea what you asked last time. We gave a LiveKit voice agent persistent memory using @MongoDB Atlas Vector Search. RAG, hybrid rankFusion recall, and a profile that loads before the agent says hello. Full walkthrough and starter kit below.
5
31
3,186
For AI avatars that feel engaged while your users are speaking, with eye contact, movements, and expressions generated live from a single reference image, check out @runwayml Characters. Add one to a LiveKit voice agent with three lines of code.
9
3
53
6,055
Ship a voice agent on any website with a single script tag. The widget supports voice, video, screen share, and text chat. Configure branding, capabilities, and per-visitor context from the LiveKit Cloud dashboard. Works on Shopify, Webflow, WordPress, or any custom site.
3
3
60
5,669
Already built a @LangChain agent? You don't have to rebuild it for voice. With the LangChain plugin for LiveKit Agents, you can connect it to a realtime voice pipeline, complete with speech-to-text, text-to-speech, and the infrastructure to deploy it at scale.
4
2
29
3,697
Your outbound phone agent has 1-2 seconds to figure out if it's talking to a person, a voicemail, or an IVR. We shipped Answering Machine Detection (AMD) in LiveKit Agents to do that for you so your agent knows when to keep talking, leave a message, use the keypad, or hang up.
4
9
112
8,471
Add a face to your voice agent. LiveAvatar by @HeyGen is now supported in LiveKit Agents. Add a realtime human avatar to your agent without rebuilding the conversation loop. Your LiveKit agent still owns the room, turn-taking, model orchestration, and voice pipeline. LiveAvatar renders the synchronized face and video stream. Useful for product demos, onboarding, tutoring, and support agents that need a visual layer.
9
7
96
11,026
Voice cloning is now available on LiveKit Inference. We’re launching with @inworld and @cartesia. Clone a voice once and use it across multiple TTS providers, with automatic fallback to the same voice if a provider fails mid-call. Free to create and available on all paid plans today.
5
9
98
11,460
We shipped structured data collection for voice agents. Use data collection mode in Agent Builder, or Tasks and TaskGroups in our Agents SDKs. Every session ends with a clean JSON payload for your CRM, form, or database. Great for lead qualification, patient intake, and surveys.
6
6
67
5,207
Build a voice agent in a day. We’re joining @MongoDB for the Agentic Evolution Hackathon in London on Saturday, May 2. Register for a chance to win £15k in prizes. The top 6 teams demo at MongoDB.local London, and the top 3 pitch live on stage. cerebralvalley.ai/e/mongo-db…
1
11
1,539
With the right prompting, Gemini 3.1 Flash TTS is one of the most expressive voice models ever. It also prompts differently than anything else we’ve tested. We built a guide on how to get the most out of it, including how to use "Director's Notes" for realism.
4
3
69
4,011
We published a guide on how to do parallel processing inside an agent session, with policy violation checks and guardrails as the example. Check out the demo and read the full post for all of the code you need to apply this pattern to your voice agents.
6
8
74
4,812
xAI STT is live. You can now run a complete cascaded voice agent pipeline on xAI (STT + Grok + TTS) through LiveKit Inference with one API key, giving you more control, full visibility, and easy component swaps.
13
12
99
10,691
We launched livekit-wakeword, an open-source library that lets you train a custom wake word model from scratch with a single command. It handles synthetic data generation, augmentation, training, and ONNX export all in one shot.
11
11
97
10,471
Load more