FAQ

What if the browser says “access to the camera/microphone is not allowed”?

Cause: the browser permission was denied, or the page isn’t in a secure context (HTTPS / localhost). Steps to fix:

Make sure the page is accessed over HTTPS or localhost (plain HTTP domains can’t access media devices)
Click the camera/microphone icon on the right side of the browser address bar and change the permission to “Allow”
If you previously chose “Always block”, reset the permission manually in the browser’s site settings:
- Chrome: ”🔒” on the left of the address bar → “Site settings” → “Camera/Microphone” → “Allow”
Refresh the page and try again

You can check in advance whether the environment can capture media through the secure and mediaDevices fields of srtc.getEnvInfo():

const env = srtc.getEnvInfo();
if (!env.secure) {
  alert('Please access this page over HTTPS or localhost');
} else if (!env.mediaDevices) {
  alert('This browser does not support media device access. Please check your permission settings');
}

What if I can receive but not publish during local development?

Cause: most likely the page is running over plain HTTP, and the browser restricts WebRTC publishing. Solution: access your local development environment via http://localhost or http://127.0.0.1, not http://[local LAN IP] (publishing isn’t supported over that protocol). For protocol and publishing support, see Integration - URL protocol restrictions.

How do I handle video/audio autoplay being blocked by the browser?

The browser’s autoplay policy forbids playing media without a user gesture; the symptom is no audio or no video after calling startPlay(). The SDK notifies you through the TRACK_AUTOPLAY_FAIL event:

srtc.onNotifyChannelEvent = (evt) => {
  if (evt.type === ChannelEventType.TRACK_AUTOPLAY_FAIL) {
    // Option 1: the SDK's built-in guide dialog (on by default)
    // Playback starts automatically after the user clicks the dialog

    // Option 2: handle it yourself and play again when the user clicks a button
    const failedTrack = evt.data; // BaseTrack
    document.querySelector('#play-btn')!.addEventListener('click', async () => {
      if (failedTrack instanceof RemoteAudioMixTrack) {
        await failedTrack.startPlay({ disableAutoPlayDialog: true });
      }
    }, { once: true });
  }
};

Best practice: call join and startPlay after the user actively clicks a “Join” or “Start call” button; this user gesture context usually gets around the autoplay restriction.

Safari 15.4 and above support getDisplayMedia (screen sharing), with these limitations:

Capturing system audio isn’t supported (getAudioTrack() returns undefined)
Screen sharing isn’t supported on iOS Safari (a system limitation)
Some older Safari versions (< 15.4) don’t support screen sharing at all

You can check in advance whether the current environment supports it with srtc.getEnvInfo().screenshare:

const env = srtc.getEnvInfo();
if (!env.screenshare) {
  alert('This browser does not support screen sharing. Please use Chrome or desktop Safari 15.4+');
}

What if joining the channel fails with TokenExpired?

The token is issued by the server and has a short validity period (usually from a few minutes to a few hours, decided by the issuer). Solution:

Get a fresh token from the server right before calling srtc.join(); don’t cache old tokens
Check that the client and server system clocks are in sync (clock skew can make the token expire early)

How do I join multiple channels at once?

Call join multiple times on one SRTC instance. Each join returns that channel’s Channel object, and each channel is operated independently:

const channel1 = await srtc.join(token1);
const channel2 = await srtc.join(token2);

await channel1.publishLocalTrack(track);
await channel2.leave(); // Leave channel 2 without affecting channel 1

For event isolation, publishing one captured track to multiple channels, and more, see Multi-channel. Note: running multiple channels in parallel uses more bandwidth and system resources, so use it only when needed.

Overview

Web SDK

Android SDK

Swift SDK (iOS/macOS)

C SDK (server/embedded)

Python SDK (server-side AI)

Server API

What if the browser says “access to the camera/microphone is not allowed”?

What if I can receive but not publish during local development?

How do I handle video/audio autoplay being blocked by the browser?

What if joining the channel fails with TokenExpired?

How do I join multiple channels at once?

​What if the browser says “access to the camera/microphone is not allowed”?

​What if I can receive but not publish during local development?

​How do I handle video/audio autoplay being blocked by the browser?

​Is screen sharing unavailable on Safari?

​What if joining the channel fails with TokenExpired?

​How do I join multiple channels at once?

What if the browser says “access to the camera/microphone is not allowed”?

What if I can receive but not publish during local development?

How do I handle video/audio autoplay being blocked by the browser?

Is screen sharing unavailable on Safari?

What if joining the channel fails with TokenExpired?

How do I join multiple channels at once?