Opus is an open, royalty-free codec (RFC 6716) that became the universal default for real-time audio and the mandatory codec in every WebRTC stack. Its trick is hybridizing two engines: a SILK speech layer from Skype and a CELT music layer, switching or blending between them per frame. That lets one codec span 6 kbps voice to 510 kbps full-band stereo, with frame sizes from 2.5 to 60 ms, so it tunes for either ultra-low latency or efficiency. Built-in in-band FEC and DTX make it resilient on lossy networks. Outside calls, Opus is also strong for streaming and podcasts where hardware support isn't required.
The book · Volume 4 of 7
Audio for Video: A Complete Guide to Sound in Video Products: From PCM and Codecs to Loudness, Accessibility, and Real-Time Voice Processing
Opus handles everything from narrowband speech to stereo music in one codec, but its many modes are easy to configure poorly. This volume covers Opus in depth with SILK and CELT modes, bitrate and frame size choices, DTX and in-band FEC, so you can tune it for calls, streams or music. Written by Nikolay Sapunov, CEO at Fora Soft.
Tuning Opus settings for voice and music quality in your WebRTC app?
Share your use cases and network conditions and we will deliver Opus settings per scenario, with listening tests and bandwidth numbers. 250+ video products shipped since 2005.
