What’s Next for Real-Time Video? Key Trends and Insights from the 2023 RTC Conference

1. Industry Trends

WebRTC 1.0 Feature Complete

Since its inception in 2011, WebRTC has matured over seven years, reaching full standardization this year, signaling a new wave of adoption.

AV1 Finalized

AV1, the first codec from the AOM Alliance, was finalized in March 2023 with Apple joining the alliance, positioning AV1 as a potential next‑generation video standard.

AI and Deep Learning Integration

Advances in AI are driving innovations in real‑time audio‑video, such as machine‑learning‑based encoding, bandwidth estimation, and real‑time image recognition.

5G Commercial Rollout

The global rollout of 5G this year is expected to spawn new RTC scenarios.

Beyond these milestones, RTC is expanding into education, healthcare, gaming, and public services, heralding a fresh wave of applications.

Education

Real‑time video enables interactive online teaching models—hand‑raising, 1v1, dual‑teacher, small‑class formats, and student expression analysis—supporting vertical services like interactive English lessons and live music instruction.

Gaming

RTC powers socially interactive game experiences such as “Werewolf” and board games, transforming traditional matchmaking into video‑enabled social play, and enabling VR/AR game experiments.

Other Use Cases

Examples include Chongqing’s 120 emergency service using RTC for pre‑arrival medical guidance, real‑time policing, securities account opening, and live quizzes.

2. Conference Topics

The two‑day event featured industry trend talks, machine‑learning and real‑time tech sessions, architecture design, and codec workshops. Highlights included:

Codec Session

High‑Quality Video Encoding Optimizations for Real‑Time Communication

Pre‑encoding: ML‑based bandwidth estimation, adaptive frame rate and resolution, content‑aware ROI encoding.

During encoding: bitrate control optimization, dynamic hardware/software switching, subjective video quality scoring.

Another focus was the self‑developed Solox audio decoder, a WebRTC‑compatible packet‑loss‑resilient codec that splits a 32 kbps stream into two complementary 16 kbps streams for improved robustness.

International Sessions

Google engineers Daniel Burnett and Huib Kleinhout presented upcoming WebRTC features, including authentication & simulcast, scalable codecs (H.265/AV1), background operation without a browser window, expanded JavaScript capabilities (media stream handling, codec parameter control), and QUIC replacing DataChannel SCTP.

Real‑Time Architecture Session

IMWeb’s VP of Technology Zhang Yan‑Jing shared how VIPKid leverages RTC for a global education platform serving over 10 million teacher clients and 150 million student clients, with a worldwide low‑latency network spanning 16 countries, 55 cities, 9 data centers, and 100 GPU servers, plus AI‑driven facial expression recognition and voice assessment.

3. Summary

Attending the RTC conference provided valuable insights into industry trends and technical advancements, especially the growing adoption of WebRTC in education and the promising future of JavaScript‑driven audio‑video processing, underscoring the need for continuous learning in this deep and evolving field.