# WebRTC Projects at AI Tinkerers > Canonical HTML: https://aitinkerers.org/technologies/webrtc > Markdown URL: https://aitinkerers.org/technologies/webrtc.md > Technology record last updated: 2026-09-18T15:13:46Z > Generated: 2026-09-21T17:51:55Z WebRTC (Web Real-Time Communication) is the open-source project and API set that enables secure, peer-to-peer (P2P) video, voice, and data exchange directly between browsers and mobile applications. WebRTC delivers high-performance real-time communication (RTC) capabilities directly to the browser, eliminating the need for plugins or proprietary software. The technology is an open standard, backed by industry leaders like Google, Apple, and Mozilla, and is implemented via three core JavaScript APIs: `MediaStream` (for accessing local camera/mic), `RTCPeerConnection` (for managing peer connections), and `RTCDataChannel` (for generic data transfer). It mandates encryption using DTLS and SRTP and handles complex networking challengesβ€”specifically Network Address Translation (NAT) traversalβ€”by leveraging STUN and TURN servers. This robust framework powers major applications, including Google Meet and Microsoft Teams, ensuring low-latency, secure P2P media streaming. - Official technology site: https://webrtc.org - Public AI Tinkerers demos and talks: 12 - Result page: 1 of 1 ## Recent Public Talks and Demos ### [The Ask Protocol](https://seattle.aitinkerers.org/talks/rsvp_6IZwJTbl1_g) I built Abject, the first abject-oriented OS built on the Ask Protocol. Chatbots give you answers that scroll away. App stores give you apps: sealed boxes, identical for everyone, deaf to each other. Abject gives you neither. An abject is a living object with a face, a memory, and opinions about its own source code. It keeps working when you close the chat. It talks to the other abjects. Some of them, it made. - Event context: AI Dev Tools Track - Seattle - August 4 β€” 2026-08-05 β€” Seattle - Public talk page: https://seattle.aitinkerers.org/talks/rsvp_6IZwJTbl1_g ### [ROSClaw: Agentic Robots with OpenClaw - "what happens when you give a robot a lobster brain?"](https://nyc.aitinkerers.org/talks/rsvp_r5hr1Aqj29U) We are presenting ROSClaw, a framework for integrating OpenClaw into ROS2 (robot operating system) compatible robots and more e.g. ESP32 devices. Winners of the SF OpenClaw Hackathon, and presenters at ClawCon 2026. We walk through installing our OpenClaw robot extension, running a set of ROS2 packages and hosting OpenClaw locally on a robot or using a hosted WebRTC signaling server. We further discuss emergent behaviors we observe that were not hard-coded into the robot(s) we have integrated on. - Event context: 🦞Demo Night: OpenClaw ft Convex β€” 2026-02-17 β€” New York City - Public talk page: https://nyc.aitinkerers.org/talks/rsvp_r5hr1Aqj29U ### [GPU-Streaming Omniverse to the Browser Per‑Org Sessions over WebRTC](https://miami.aitinkerers.org/talks/rsvp_Mq1fLtz3A-E) Live build walkthrough of a prototype that spins up an isolated NVIDIA Omniverse/Kit session per organization, then streams it into a standard web browser via WebRTC. I’ll show the end-to-end flow: request β†’ provision Azure GPU instance β†’ boot the Kit runtime β†’ establish a WebRTC connection β†’ interact in-scene β†’ teardown/cleanup. Focus is on the state machine, orchestration, and the practical failure modes (startup latency, ICE/networking, cleanup). - Event context: AI Tinkerers Miami Meetup β€” The 2026 Builder Kickoff β€” 2026-02-11 β€” Miami - Public talk page: https://miami.aitinkerers.org/talks/rsvp_Mq1fLtz3A-E ### [Tambourine: Your personal voice interface for any app](https://singapore.aitinkerers.org/talks/rsvp_GcrcHQwpqik) Tambourine is an AI-powered voice dictation tool that can be used with any text box. A fully open source alternative to Wispr Flow. It allows users to fully own the AI pipeline by swapping models and defining custom formatting logic. I will demo how to use it and talk about what situations AI voice dictation does particularly well in. I will also go into the technical details of how this was built on top of other open source frameworks, Tauri and Pipecat, as well as the infrastructure powering real-time voice AI like WebRTC. - Event context: AI Tinkerers - The Age of AI & Infrastructure (Singapore) β€” 2026-02-11 β€” Singapore - Public talk page: https://singapore.aitinkerers.org/talks/rsvp_GcrcHQwpqik ### [DIY Wedding Translator for Three Languages](https://tokyo.aitinkerers.org/talks/rsvp_YG0dq12PBoM) I recently hosted a wedding ceremony in Tokyo and needed live captions in English, Mandarin, and Japanese so every guest could laugh or "cry" together. Hiring a professional translator who speaks all three languages was almost unrealistic and too expensive, so I hacked together a browser-based translator that listens, slices audio on a steady beat, and beams subtitles to a shared screen. I’ll demo the heartbeat timer, the playful operator dashboard, and the safety checks that kept the captions flowing while I emceed. - Event context: AI Tinkerers Tokyo Kickoff – October 10, 2025 β€” 2025-10-10 β€” Tokyo - Public talk page: https://tokyo.aitinkerers.org/talks/rsvp_YG0dq12PBoM ### [Making an esp32 microcontroller conversational with WebRTC](https://nyc.aitinkerers.org/talks/rsvp_9YDD7S3E_gI) This talk walks through the journey of bringing conversational AI to one of the smallest devices yet β€” an ESP32 microcontroller. Using an M5Stack AtomS3R and Atomic Echo Base, I integrated my ElevenLabs voice clone with the Pipecat AI framework and WebRTC for real-time, low-latency audio streaming. The talk covers the challenges of working with constrained hardware, why LLMs aren’t a magic bullet here, and how open-source contributions from the Pipecat community made this possible. Attendees will see how a tiny embedded device can become a fully conversational voice agent. - Event context: AI Tinkerers August Demo Day ft. Auth0 and Comet. Demo what you've built! β€” 2025-08-26 β€” New York City - Public talk page: https://nyc.aitinkerers.org/talks/rsvp_9YDD7S3E_gI ### [HRMNY HR: Employee Dispute Resolution Voice Agent Platform](https://nyc.aitinkerers.org/talks/rsvp_b-TU7YzqpIQ) 1.) I will walk through the code of the platform I have built. 2.) Mistakes I have made, corners I have cut and the things that have worked well 3.) How I use LLMs : specifically How/Why I used realtime-preview models for the Voice Agent and not used the traditional models that need STT and TTS. 4.) I use multiple LLMs in my product: 1 for Voice Agent (as stated above), 1 for Analyzing transcripts and 1 for generating reports that need to face HRs Each Agent comes with its own set of instructions. And I can talk about how I manage (code-wise) using the LLMs. I am right now exploring how to make the LLMs perform better and I can share those findings as I get them in the next few days - Event context: AI Tinkerers August Demo Day ft. Auth0 and Comet. Demo what you've built! β€” 2025-08-26 β€” New York City - Public talk page: https://nyc.aitinkerers.org/talks/rsvp_b-TU7YzqpIQ ### [Barveen kumar - Co-Founder](https://tiruchirappalli.aitinkerers.org/talks/rsvp_DqzU5-ALd9U) Founder of KuralynX, building AI voice agents with contextual and affective computing capabilities. We replace traditional IVR systems with deployable voice interfaces that handle real-time conversations, capture intent beyond keywords, and reduce business losses from unanswered or mishandled calls. - Event context: πŸ“£ Join us for our First AI Tinkerers Trichy Meetup πŸ“£ β€” 2025-07-12 β€” Tiruchirappalli - Public talk page: https://tiruchirappalli.aitinkerers.org/talks/rsvp_DqzU5-ALd9U ### [Tayle: Building an AI-Powered Voice Interview System for Life Story Capture (using OpenAI's Realtime API) + Open-Source SDK](https://seattle.aitinkerers.org/talks/rsvp_eGZLeyw02QI) We're building Tayle, an AI-powered platform for capturing and sharing life stories through natural voice interviews. We'll demo our voice UX, showing how we leverage OpenAI's Realtime API and LLMs to guide conversations, handle interruptions, and extract key points into a coherent engaging story worth sharing with friends and family. We've also open-sourced a TypeScript SDK for working with the Realtime API, complete with a working example and WebRTC integration. - Event context: AI Tinkerers Seattle - February 2025 Meetup β€” 2025-02-22 β€” Seattle - Public talk page: https://seattle.aitinkerers.org/talks/rsvp_eGZLeyw02QI ### [Revoice Live - online voice changing](https://berlin.aitinkerers.org/talks/rsvp_VS9p-7oUAQ4) The talk covers the design and development of an online voice-changing application. Using voice conversion as an example, we will look into how to build real-time sequence processing systems. - Event context: AI Tinkerers Berlin - August 22 β€” 2024-08-22 β€” Berlin - Public talk page: https://berlin.aitinkerers.org/talks/rsvp_VS9p-7oUAQ4 ### [Combining Multiple Public Pre-Trained Models for Better Audio Segmentation](https://toronto.aitinkerers.org/talks/rsvp_t6jOg5jQ84M) The work-in-progress includes an audio processing pipeline. The audio is segmented and transcribed. Linguistic text features are extracted from the transcribed text. The use of publicly pre-trained AI models has provided high quality results. The next development stage will be focused on finer segmentation of the audio using the outputs of these public models. - Event context: AI Tinkerers Toronto - Inaugural Meetup β€” 2024-04-11 β€” Toronto - Public talk page: https://toronto.aitinkerers.org/talks/rsvp_t6jOg5jQ84M ### [Federated Learning with Flame](https://seattle.aitinkerers.org/talks/rsvp_HniMKJNpKf8) Traditional machine learning depends on the centralization of data, but that comes with privacy and computational concerns. A reality with billions of edge devices diminish those issues, especially with the advent of federated machine learning. Training may be performed on edge devices directly, keeping datasets decentralized and private. Additionally, offloading work to different nodes means less computation per device. Projects like GBoard, Siri, and even the medical and military fields already use federated learning. One current open-source framework for federated machine learning is Project Flame (maintained by Cisco Systems). Flame uses object-oriented programming to implement different graphs between edge devices for a federated learning network. Flame can be extended to different kinds of topologies and executed across multiple devices that run Python code using P2P communication. - Event context: AI Tinkerers Seattle - February 2024 Meetup β€” 2024-02-29 β€” Seattle - Public talk page: https://seattle.aitinkerers.org/talks/rsvp_HniMKJNpKf8 ## Related Technologies - [TypeScript](https://aitinkerers.org/technologies/typescript) ([Markdown](https://aitinkerers.org/technologies/typescript.md)) β€” 205 public demos - [ESP32](https://aitinkerers.org/technologies/esp32) ([Markdown](https://aitinkerers.org/technologies/esp32.md)) β€” 2 public demos - [OpenAI](https://aitinkerers.org/technologies/openai) ([Markdown](https://aitinkerers.org/technologies/openai.md)) β€” 112 public demos - [Pipecat](https://aitinkerers.org/technologies/pipecat) ([Markdown](https://aitinkerers.org/technologies/pipecat.md)) β€” 7 public demos - [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) β€” 662 public demos - [SIP](https://aitinkerers.org/technologies/sip) ([Markdown](https://aitinkerers.org/technologies/sip.md)) β€” 2 public demos - [Azure GPU VMs](https://aitinkerers.org/technologies/azure-gpu-vms) ([Markdown](https://aitinkerers.org/technologies/azure-gpu-vms.md)) β€” 1 public demo - [Bandwidth](https://aitinkerers.org/technologies/bandwidth) ([Markdown](https://aitinkerers.org/technologies/bandwidth.md)) β€” 1 public demo - [BitTorrent](https://aitinkerers.org/technologies/bittorrent) ([Markdown](https://aitinkerers.org/technologies/bittorrent.md)) β€” 1 public demo - [C++](https://aitinkerers.org/technologies/c) ([Markdown](https://aitinkerers.org/technologies/c.md)) β€” 10 public demos - [Containerization](https://aitinkerers.org/technologies/containerization) ([Markdown](https://aitinkerers.org/technologies/containerization.md)) β€” 3 public demos - [Docker](https://aitinkerers.org/technologies/docker) ([Markdown](https://aitinkerers.org/technologies/docker.md)) β€” 147 public demos - [Edge computing](https://aitinkerers.org/technologies/edge-computing) ([Markdown](https://aitinkerers.org/technologies/edge-computing.md)) β€” 6 public demos - [ElevenLabs](https://aitinkerers.org/technologies/elevenlabs) ([Markdown](https://aitinkerers.org/technologies/elevenlabs.md)) β€” 42 public demos - [Federated learning](https://aitinkerers.org/technologies/federated-learning) ([Markdown](https://aitinkerers.org/technologies/federated-learning.md)) β€” 2 public demos - [G](https://aitinkerers.org/technologies/g) ([Markdown](https://aitinkerers.org/technologies/g.md)) β€” 1 public demo - [Google](https://aitinkerers.org/technologies/google) ([Markdown](https://aitinkerers.org/technologies/google.md)) β€” 9 public demos - [Grafana](https://aitinkerers.org/technologies/grafana) ([Markdown](https://aitinkerers.org/technologies/grafana.md)) β€” 3 public demos