iRTE2026Get your pass

The program / iRTE 2026

Find your
next breakthrough.

Explore 14 tracks across two days of technical deep dives, industry perspectives and practical ideas. All times are Beijing time (UTC+8).

Program subject to change. Session titles marked “tentative” are provisional.

OCT 23

Opening keynotes

The next chapter of real-time intelligence

Meet the people shaping the next wave of real-time interaction, from conversational AI to connected devices.

Conference program

View 4 sessions

  • Talking to AGI: The next chapter of real-time interaction Tentative title

    Tony Zhao · Founder & CEO, Shengwang

  • Models, automation—what comes next? Tentative title

    Yangqing Jia · Founder & CEO, Intent Lab

  • Panel discussion

    Tony Wang (Moderator) · Co-founder, Agora

  • Panel discussion

    Bo Cui (Moderator) · Chief Business Officer, Shengwang
    Wei Chen · Founder, Xieyue Intelligence
    Zhili Liu · CEO, Beijing Guanke Technology
    Guoxiang Cai · Co-founder & Head of Ecosystem, Rokid
    Kan Wu · CEO, Titan Qiming

OCT 23

Shengwang product showcase

What’s next from Shengwang

Explore the latest product releases and engineering advances across real-time communication and AI.

Curated by Guanghua Yao and Ting Li. Media partner: Woshipm.

View 11 sessions

  • The latest in AI at Shengwang

    Guanghua Yao · Head of AI Products, Shengwang

  • RTE Next: A unified communication foundation for real-time agents Tentative title

    Xuebin Yang · Shengwang

  • From conversation to outcomes: Voice infrastructure for enterprise agents Tentative title

    Yujie Mao · Shengwang

  • AI + RTE in practice: An international perspective Tentative title

    Ben Weekes · Agora

  • IM 5.0: The next generation of intelligent messaging Tentative title

    Can Wang · Shengwang

  • Real-time transcription and translation at scale

    Fan Yang · Shengwang

  • A reusable engine for real-time meetings and collaboration Tentative title

    Xiangzhong Ji · Shengwang

  • Beyond recording: Connecting media in the cloud Tentative title

    Rongjun Mei · Shengwang

  • Licensed karaoke content and beauty effects for richer entertainment Tentative title

    Shihao Rao · Shengwang

  • Security and compliance behind real-time intelligence Tentative title

    Yuan Xue · Shengwang

  • Real-time intelligence for good Tentative title

    Lei Qian · Shengwang

OCT 23

Enterprise voice agents

From conversation to completion

What does it take for a voice agent to get the job done? Explore the technology, workflows and business metrics behind enterprise customer service agents that deliver measurable results.

Curated by Fan Yang and Ying Li. Media partner: Jazzyear.

View 5 sessions

  • Voice agents that communicate and convert: Shengwang’s RMB 2,999-per-month offering

    Zhe Huang · Call Agent Product Manager, Shengwang

  • Finding opportunity in the next wave of AI voice agents

    Zhili Liu · CEO, Beijing Guanke Technology

  • Enterprise voice agents: Technology and practice

    Jun Xu · Senior Technology Expert, Meituan

  • Building customer service agents that keep improving

    Xinhua Shi · Senior Architect, Alibaba ATH

  • From cascaded pipelines to end-to-end voice agents: The opportunity for ASR and TTS

    Ruhuan Zhang · Senior AI Architect, MiniMax

OCT 23

Conversational AI

Intelligence through interaction

From end-to-end speech models to systems that listen, speak, reason and act at once, explore the architectures and full-duplex interactions shaping the next generation of conversational AI.

Curated by Lei Xie, Northwestern Polytechnical University, and Bo Wu, Shengwang.

View 5 sessions

  • Audio intelligence in the foundation model era: From understanding and generation to interaction and agents

    Lei Xie · Professor & Director, Audio, Speech and Language Processing Lab, Northwestern Polytechnical University

  • Multimodal models and real-time voice interaction Tentative title

    Zeyuan Wang · Audio Algorithm Engineer, Shengwang

  • Target speaker extraction and multi-speaker recognition in complex conversations

    Ming Li · Professor, CUHK-Shenzhen

  • Loxi-Speech: Speech synthesis built for real-time interaction

    Xinsheng Wang · Head of Loxi, Soul; Co-founder, Loxinity

  • Panel: Bringing real-time interaction models into production

    Bo Wu (Moderator) · Head of Multimodal AI, Shengwang
    Guangjian Xu · Head of Audio Technology, Liangyuan Xinchuang
    Zhifei Xie · PhD, NTU; Creator of VoiceMem, Mega-ASR and Mini-Omni
    Xinsheng Wang · Head of Loxi, Soul; Co-founder, Loxinity

OCT 23

RTE 101

Your first production-ready voice agent

Bridge the gap between a promising demo and a voice agent people can use. Get grounded in audio pipelines, full-duplex communication, interruption handling and latency optimization.

Curated by Huancheng Bai, RTE Developer Community, and Zhen Chen, Agora.

Session details to be announced.

OCT 23

Demo Day

Meet the next wave of real-time experiences

Discover emerging conversational AI applications and connected devices. See what builders are experimenting with, challenge familiar assumptions and find inspiration for your next project.

Presented by the Supersonic Program and RTE Developer Community.

Session details to be announced.

October 23–24 · Beijing

Be part of
what happens next.

Two days of ideas, working demos and conversations with the people building real-time intelligence.

Get your pass