The first audio and video technology conference of Netease ended in 2021。 Technical experts from many audio and video fields, such as Netease smart enterprise, Netease game, Netease cloud music, Netease Hangzhou Research Institute and so on, gathered together, combined with their own research and experience for many years, to communicate with the audienceWe have jointly discussed the cutting-edge technological innovation of audio and video and shared the practical achievements of audio and video applications, which has brought many new thoughts and opinions to the development of the industry.
As we all know, vision is the main source of information for the perception of all things. Hearing makes the silent information further and become “vivid”. In the era of mobile Internet, online learning, work and entertainment with “good voice and color” are widely popular, and new scenes such as interactive live broadcasting, video conference and remote recruitment have emerged one after another. Behind it is the strong support of audio and video technology, which imperceptibly integrates the online scenes of all walks of life.
Based on this background, the first audio and video technology conference of Netease came into being.Closely focusing on “color” and “sound”, the conference set up two special chapters: “wide vision: vision of video technology innovation” and “sound in its environment: audio technology immersion experience”, the whole process lasted for two days, and a hearty science and technology feast was dedicated to all guests and online audiences.
Netease audio and video technology conference
Develop technology and promote exchanges
First of all, Dr. Chen Gong, the general producer of this conference and VP of Netease intelligent enterprise technology, described the opportunity and purpose of this conference. Chen Gong said that in recent years, audio and video technology has developed rapidly and commercialization has accelerated, while Netease continues to make technological innovation and breakthroughs in this field andAt the beginning of this year, Netease tm599 audio and video technology sub committee was established, and it is expected that the sub committee will focus on the accumulation of technical capabilities of Netease group in various segments of audio and video, and promote exchanges and cooperation in the industry.
Netease intelligent enterprise technology VP Chen Gong
Vision of video technology innovation
At the special session of “wide vision: vision of video technology innovation”, Han Qingrui, the producer of the special session and senior technical expert of Netease audio and video laboratory, introduced the main contents of the special session. Combined with the practical experience of Netease cloud music, Netease shield, Netease Yunxin and Netease mutual Entertainment in video technology, this session will focus onVideo processing, video deep forgery detection, content security, video enhancement and computer vision technology, AI dance synthesisAnd other topics.
Han Qingrui, senior technical expert of Netease audio and video Laboratory
Sui Shichen, senior video algorithm engineer of Netease cloud music, cut into the hot short video field from a technical perspectiveThis paper deeply analyzes the functions and design ideas of various video creation tools in cloud musicIn the end, technology is only a tool to assist the creator to express value, rather than directly replace the creator’s ideas.
Sui Shichen, senior video algorithm engineer of Netease cloud music
Hu Yifeng, senior image algorithm engineer of Netease Yidun, introducedThe “double-edged sword” effect of AI in various scenesTaking the most prominent hidden danger of face forgery as an example,From the perspective of algorithm and application, this paper shares Netease shield’s solution strategies and remarkable achievements in video deep forgery detection.
Hu Yifeng, senior image algorithm engineer of Netease Yidun
Zhou Chenhui, senior video algorithm engineer of Netease Yunxin, sorted out the causes of a series of problems affecting video definition, color and quality, and shared solutions based on AI video processing algorithm one by one,Netease Yunxin image quality enhancement has great potential in real-time audio and video interaction, low delay live broadcast, on-demand and other scenes.
Zhou Chenhui, senior video algorithm engineer of Netease Yunxin
Tan Zhipeng, senior AI research engineer of Netease mutual entertainment, focused on the dance animation commonly seen in the game and film and television industries, and explained the difficult process behind the generation of dance animation, as well asHow to produce high-quality dance animation quickly and efficiently through AI music and dance synthesis algorithm technology.
Netease mutual entertainment senior AI research engineer Tan Zhipeng
Sound is in its territory
Audio technology immersion experience
The special session of “sound in its environment: audio technology immersion experience” is also full of technical dry goods. The novel coronavirus pneumonia, a special producer and Liu Huaping, director of NetEase cloud music audio and video lab, has greatly promoted the development of online application scenarios.Real time audio and video call is a very “just needed” technical point in many applications, and the sound quality is also one of the most core parameters in the audio and video call system.
Liu Huaping, head of Netease cloud music audio and video Laboratory
Hao Yiya, an expert of Netease Yunxin audio algorithm, first reviewed the background and main application scenarios of RTC real-time communication, and introducedArchitecture and effect of Netease Yunxin AI noise reduction technology, and echo cancellation algorithm. In terms of the construction of audio standardization evaluation system, Netease Yunxin has also made a lot of efforts, such as setting up an audio laboratory and establishing a noise reduction algorithm evaluation system, and looks forward to continuously promoting the development of domestic RTC audio field. Finally, I shared other audio capabilities such as 3D sound effect and AEC of Netease Yunxin and the academic research results of audio laboratory.
Hao Yiya, Netease Yunxin Audio Algorithm expert
Zhao Xiangyu, head of Netease cloud music audio and video algorithm, said,Immersive audio is very important for users to get a real and immersive experience。 Around this theme, Zhao Xiangyu listed the main factors affecting the sense of immersion in the sound field and the technical solutions. Finally, the audio effect of the technical scheme is displayed, which makes all the on-site and online viewers “immersed” together.
Zhao Xiangyu, head of Netease cloud music audio and video algorithm
Liu Dong, Yang Zhen and Li Xiang, speech algorithm experts from Netease Hangzhou Research Institute, have focused on the R & D and application of AI technology in speech related fields for many years, and from their respective deep cultivationAudio understanding system, acoustic model in speech recognition, online reasoning system of speech recognitionThe three dimensions describe the technical challenges, solutions and practical cases faced in the R & D process.
Liu Dong, Yang Zhen and Li Xiang, speech algorithm experts of Netease Hangzhou Research Institute
Fu Mingming, a Netease game thunder fire audio design expert, shared the theme of the development and application of AI music. Fu Mingming proposed that AI music is a cross field of algorithm and art. Its essence is to analyze and learn music data through various algorithms to form a relatively aesthetic style model, and generate content in the selected style model based on user input.
Fu Mingming, Netease game thunder fire audio design expert
The conference has ended
Technology sharing never stops
The first audio and video technology conference of Netease in 2021 has been successfully concluded. At the conference, wonderful speeches from experts in various segments of audio and video,It not only provides practical technical solutions for industry practice and application, but also provides new thoughts and opinions for industry development, and guides the future technical direction and development trend.
The conference has ended and technology sharing has never stopped。 The insights of the lecturers attracted more than 54000 viewers and were widely recognized by the audience. To facilitate the audience to review and promote the sharing of technology, the speech video will be released in the salon, cloud commerce, MCtalk, Bilibili and other platforms. The relevant content will also be released in the NetEase WeChat technology + official account.
Sweep the official account and get the latest information.