Synton Developer Guides

Step-by-step instructions for integrating the browser WebRTC SDK, loading standalone Opus and VP9 WebAssembly decoders, and requesting dedicated tenant provisioning.

1. Embeddable Web SDK Quickstart

Your backend issues a short-lived room token scoped to the caller and room name. The browser client loads sfu-client.js (livekit-client v2.22.3, Apache-2.0) and connects to wss://rtc.synton.cc.

quickstart.html
<script src="../assets/v1/client/sfu-client.js"></script>
<script>
  async function joinStandup(roomToken) {
    const room = new LivekitClient.Room({
      adaptiveStream: true,
      dynacast: true,
      videoCaptureDefaults: { resolution: LivekitClient.VideoPresets.h720.resolution }
    });

    room.on(LivekitClient.RoomEvent.TrackSubscribed, (track) => {
      document.getElementById("stage").appendChild(track.attach());
    });

    await room.connect("wss://rtc.synton.cc", roomToken);
    await room.localParticipant.enableCameraAndMicrophone();
  }
</script>

2. Standalone WASM Audio & Video Codecs

sfu-client.js uses native browser WebRTC pipelines for live calls. Separately, Synton distributes two standalone WebAssembly decoder packages from ogv v1.9.0 (MIT) for software decoding pipelines: codec-audio.wasm (Opus 1.3.1 via libopus) and codec-video.wasm (VP9 via libvpx). Each module includes its matching Emscripten loader.

Manifest JSON Third-Party Notices
load-wasm-codecs.html
<script src="../assets/v1/client/codec-audio-loader.js"></script>
<script src="../assets/v1/client/codec-video-loader.js"></script>
<script>
  const opus = await OGVDecoderAudioOpusW({
    locateFile: () => "../assets/v1/client/codec-audio.wasm"
  });
  const vp9 = await OGVDecoderVideoVP9W({
    locateFile: () => "../assets/v1/client/codec-video.wasm"
  });
  opus._ogv_audio_decoder_init();
  vp9._ogv_video_decoder_init();
</script>
Artifact Role Upstream Release Size SHA-256
sfu-client.js webrtc-sfu-sdk livekit-client v2.22.3 594,114 B 7fa17e37af5e996d8a25f15a637dcc0620215bc01b394e5d209f726afe7dc04d
codec-audio.wasm wasm-audio-decoder ogv v1.9.0 154,322 B af1acdfa07f389d261d13b5a6b58d8dc44e6d9dff1265ae760ae8b1470cc5fa0
codec-audio-loader.js wasm-audio-loader ogv v1.9.0 7,270 B b250fd51637998eb03832f14d07932da3cc978bd80b0199ae26b038154421baa
codec-video.wasm wasm-video-decoder ogv v1.9.0 230,038 B 79efca8f980458be2abadfc0531a17036326685102ba281f172e2c9eb683bdcb
codec-video-loader.js wasm-video-loader ogv v1.9.0 8,738 B 012e2daaa34fa84d53520ab8bfa79a950aa922a2663405f4ddaa5f83cecaf435

3. Core Deployment Scenarios

01 / STANDUPS

20-Participant Video Standups

Enable 3-layer simulcast (180p, 360p, 720p) on publishers and dynacast: true so unrendered speaker feeds pause upstream automatically.

02 / NAT FALLBACK

Enterprise Symmetric NAT Fallback

Leave iceTransportPolicy: "all" active so users behind strict corporate egress seamlessly fall back to regional TCP/TURN relays without dropping audio.

4. Developer FAQ

  • How do Adaptive Stream and Dynacast avoid saturating subscriber downlinks in crowded rooms?

    Adaptive Stream monitors the rendered pixel size and visibility of each subscribed video element and instructs the SFU to deliver only the lowest sufficient simulcast rung (180p for thumbnails, 360p for grid tiles, 720p for active focus). When no subscriber displays a track, Dynacast pauses that encoder layer at the publisher.

  • What happens when a firewall blocks UDP media and triggers fallback signaling?

    WebRTC ICE probing detects unreachable UDP candidates and pivots to regional TCP/TURN relay endpoints. Voice tracks retain priority queuing so conversation latency stays bounded even while video scales down to lower simulcast layers.

  • Why do signaling tickets expire and get issued by the application backend before joining?

    Issuing short-lived room-scoped Bearer JWTs on your own server keeps signing secrets off client devices and binds each token to one participant identity and room window. Missing or expired tokens fail immediately with HTTP 401/403 during the /rtc WebSocket upgrade.

5. Dedicated Tenant Provisioning & Enterprise Inquiry

Self-service registration is closed. Dedicated single-tenant capacity—including dedicated WebRTC media forwarding pipelines, per-project API credentials, and isolated recording egress provisioned per workspace—is allocated under enterprise agreements.

Direct Email: hello@synton.cc