Distributed audio and video analysis for real-time web communication

用于实时网络通信的分布式音频和视频分析

• 批准号: 538879-2019
• 负责人: Shea, Ryan
• 金额: $ 1.82万
• 依托单位: Simon Fraser University
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2019
• 资助国家: 加拿大
• 起止时间: 2019-01-01 至 2020-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=680600
• 关键词: Distributed; audio; video; analysis; real

Frozen Mountain Software, a British Columbia based company, provides solutions for real-time audio and video streaming to customers in over 51 countries. Through the use of Web Real Time Communication (WebRTC) based services and software development kits (SDK) Frozen Mountain enables organizations to incorporate advanced features such as WebRTC streaming, screen sharing and shared whiteboards into their applications. Recent developments in distributed audio and video analysis has led to the ability to perform the complex task of deep learning on audio and video. Through this collaboration researchers at Simon Fraser University along with developers at Frozen Mountain will develop novel methods to perform distributed analysis on video and audio streams while preserving the stringent latency constraints of real-time communication. The outcome of this collaboration will develop technology with the ability to perform complex analysis of video and audio streams to facilitate the creation of a suite of tools with use cases such as real-time language translation and sentiment analysis.

Frozen Mountain Software是一家总部位于不列颠哥伦比亚省的公司，为超过51个国家的客户提供实时音频和视频流解决方案。通过使用基于Web真实的实时通信（WebRTC）的服务和软件开发工具包（SDK），Frozen Mountain使组织能够将WebRTC流媒体、屏幕共享和共享白板等高级功能纳入其应用程序。分布式音频和视频分析的最新发展使得能够在音频和视频上执行复杂的深度学习任务。通过这项合作，西蒙弗雷泽大学的研究人员沿着Frozen Mountain的开发人员将开发新的方法来对视频和音频流进行分布式分析，同时保留实时通信的严格延迟限制。此次合作的成果将开发能够对视频和音频流进行复杂分析的技术，以促进创建一套具有实时语言翻译和情感分析等用例的工具。