Released by IDC, an international authoritative data company
"China Video Cloud Market Tracking (2023 H1)" Report
Since 2018, Alibaba Cloud has continued to maintain
China's video cloud overall market ranks first
The overall market share reached 24.4%
01 Beyond the first, above the trough
Recently, the international authoritative data company IDC recently released the "China Video Cloud Market Tracking".In the first half of 2023, the scale of China's video cloud market will reach 46.2billion dollars. According to IDC definition, the data covers all public cloud and dedicated cloud infrastructure for video application scenarios, audio and video content distribution network CDN, and video products/services deployed on video cloud infrastructure / solution, of which China’s video cloud infrastructure market size in the first half of 2023 is US$3.66 billion, and the video cloud solution market size is US$960 million.
Data shows that in the first half of 2023, Alibaba Cloud Video Cloud's overall market share reached 24.4%, and it still steadily maintains the first place in China's overall video cloud market. This is also the continuous first place for Alibaba Cloud Video Cloud since IDC released data in 2018. .
As a leader in the track for many years, Alibaba Cloud Video Cloud has witnessed the changes in the life cycle of the track, from the first year of live broadcasting to the increase in the epidemic, from the post-epidemic era to the metaverse, from cost reduction and efficiency improvement to VR phantom. Just as through the data in this IDC report, we can get a glimpse of the short-term weakness of the track.
The report shows that in the first half of 2023, the size of China's video cloud market fell by 7% year-on-year. This was also affected by multiple factors such as the slowdown in end-user entertainment demand after the epidemic and the reduction of IT expenditures by industry customers. Compared with the same period in 2022, the video cloud infrastructure Both Markets and Solutions Markets also showed slight declines.
Of course, the flight speed of a track will always change rapidly. Just as IDC analysts said, "2023, as The chill brought about by the weak growth of major customers in the past continues to be transmitted upstream.The video cloud market is experiencing the trough of the past decade and possibly the next decade.”
We have indeed witnessed the cyclical adjustments to China's video cloud market due to the objective environment, but we can also see the flickering light caused by GPT4.0 to the digital intelligence of the industry, and the explosive potential of AI brewing in the video cloud track.
It is conceivable that after the trough, there must be a foreseeable upward trend, which is also the next perspective we turn to.
02 Video Cloud 2.0 in progress
Following the backburn of AI, the audio and video field may no longer be a rough and rude way to grow with the wind, but a more valuable application exploration. Its applications are wider and deeper, giving new life to the video cloud track. This is also Alibaba Cloud Video Cloud focuses on the forward direction.
Therefore, AI has upgraded the video cloud to 2.0, making the field of view wider.
From Alibaba Cloud's perspective, the core of the video cloud is evolving: If the cloud has given video and brought about Video Cloud 1.0, then now it is AI that has given the video cloud and brought about Video Cloud 2.0. 1.0 is transitioning to 2.0.
In the 1.0 era, video applications mainly focused on the consumer Internet field, and the core solution was cost and experience issues, that is, audio and video enterprise costs on the B-side and user experience on the C-side. With the rise of AI computing power and large models, in the 2.0 era, video applications have turned to the industrial Internet field to solve the problem of scenario-based audio and video intelligent applications in various industries.
Going further, Video Cloud 2.0 is helping the digital intelligence of the industry through basic large models and small scene models.
In this regard, Alibaba Cloud Video Cloud has three key strengths:Scale strength. After years of accumulation in Video Cloud 1.0, Alibaba Cloud has become a global and one of the largest video cloud infrastructure service providers in China, providing high-quality, cost-effective video services;Intelligence, based on Alibaba Cloud Tongyi A large model that provides intelligent product upgrades across the entire chain of media asset production, processing, transmission, and consumption; scenario power based on diversified industry practices , provides a highly easy-to-use, high-performance video terminal service suite, comprehensively lowering the threshold for customer use, and at the same time, it can collaborate with scene-based small models to realize the flourishing of industry applications.
From digitization to digital intelligence, the deep integration of video cloud and AI, and in-depth exploration of industries and scenarios, are where the long-term value lies.
IDC report shows that,Aiming at the video cloud solution market with a market size of US$960 million in 2023H1,the dynamics of the two industries are obviously, one is The demand for radio and television media is growing steadily, and one is that the demand in the education industry is effectively recovering.
The Radio and Television Media industry is highly consistent with the audio and video field, and Alibaba Cloud Video Cloud has been strengthening its "intelligence" by penetrating its entire link.
After several years of development,“People’s Daily AI Editorial Department” is a typical example.
This was jointly developed by Alibaba Cloud Video Cloud and the People's Daily Newspaper Technology Department in 2020, and has played an important role in the coverage of multiple sessions of the two sessions. Since its evolution, the AI editorial department of People's Daily has iterated to version 4.0, which can now realize automated, batch and intelligent production of various new media products. Especially recently, this product won the first prize of the "King's Choice Journalism Science and Technology Award" asthe only cross-media technology award in the news industry approved by the National Science and Technology Award Office, used to commend outstanding contributions to the advancement of news science and technology, which to a certain extent reflects the power of Video Cloud 2.0 to promote the industry.
Looking at education, if it is not limited to online education, Video Cloud 2.0 can be a reshaping oftraditional education. Among them, the digital and intelligent innovation of art examinations can be called a benchmark that illuminates the industry.
In 2023, Alibaba Cloud Video Cloud cooperated with the China Academy of Art to innovate, based on the intelligent remote proctoring PaaS+ solution, the first "cloud art exam" new model, ensuring the smooth online art exam for 40,000+ candidates at home and abroad , at the same time, based on the high complexity of various art examination scenarios, we innovatively develop "intelligent anti-cheating" capabilities to maximize educational fairness while ensuring the efficiency of institutions. The relevant person in charge of information technology at the Academy of Fine Arts vividly described this as"a revolution in the history of examinations over a thousand years", which integrates AI technology on top of today's audio and video digital drive , to help more art schools adopt inclusive applications. This year, the solution also won the Benchmark Award and First Prize in the Digital Transformation Application Contest of the China Academy of Information and Communications Technology’s "Dingxin Cup" Awardhonor.
From the perspective of art colleges and universities, the "online art exam" is not just a forced adoption in response to the epidemic. It is the general trend in the development of art major admissions exams. Video Cloud happens to have the ability and opportunity to participate in the reshaping of education.
In addition, Alibaba Cloud Video Cloud is also promoting upgrades, such as remote quality inspection, telemedicine and other fields. It can be seen that the video cloud superimposed with AI no longer satisfies the consumer Internet waterfall in one dimension. It does not simply focus on live broadcast extension or on-demand innovation, but also penetrates into the TOB industry field with complex textures. Do something more difficult to explore and more socially meaningful, and this is where the creativity of Video Cloud 2.0 takes place.
03 Large model and fully intelligent Pathfinder
At the recent Yunqi Conference, Wang Jian, founder of Alibaba Cloud, predicted that "the combination of artificial intelligence and cloud computing will bring about the third wave of cloud computing."
2023 is probably the first year of the big model, with the centralized explosion of artificial intelligence and cloud computing. The GPT model has made computers have a very important impact on the revolution of technological innovation. The latest OpenAI Developer Conference will move the APP Store era to the GPT Store era.
The power of artificial intelligence and large models is creating enthusiasm and overcoming obstacles. Alibaba Damo Academy has recently achieved a great breakthrough in cancer screening using "CT scan + AI". The combination of cloud computing and AI has given society a A different kind of warmth.
With the help of cloud + AI, video cloud 2.0 can also do more. Alibaba Cloud Video Cloud is also a pathfinder for fully intelligent layout and large video cloud model.
In the past year, the video cloud team has dived into the field of large models. Based on the characteristics of the audio and video business, it has designed and built a system architecture for video cloud large model algorithm development.The entire The system covers the entire link of analysis, planning, reasoning, evaluation, training and fine-tuning, and the system is evolvable and decision-making. Among them, multiple algorithm atoms were developed, involving multiple links in the entire link of audio and video production, processing, management, transmission and distribution, playback and consumption.
For example, in the audio and video production process, Alibaba Cloud Video Cloud has developed multiple large model-based algorithms such as real-life cutout, sound cloning, text-based mapping, image-based mapping, and AI composition. Among them, the human voice cloning ability has been in-depth polished by the algorithm, making the cloned voice almost indistinguishable from the original voice of the character. Furthermore, combined with voice-driven digital human technology, human voice cloning can also create highly realistic and natural virtual digital humans. This product is now online and has attracted widespread attention.
At the same time, considering that relying solely on large models can achieve some core basic functions, but there is still a long way to go before "solving it well", the video cloud technology team also proposed the collaboration of several large and small models, that is, allowing large and small models to interact with each other. Cooperate and leverage their respective strengths.
Not only that, we know that large models provide a variety of problem-solving tools, such as question and answer, dialogue, graphs, video descriptions, etc. These tools are constantly being improved and their capabilities are getting stronger, but basically They are all solving one-dimensional problems. Therefore, Alibaba Cloud Video Cloud is also exploring large models with sensing, planning, and action capabilities.
On top of the large model, it is for better applications.
In recent years, in order to pursue the ultimate balance of quality, cost, computing power, performance and other multi-dimensional dimensions, Alibaba Cloud Video Cloud has invested heavily infull-link and full-intelligence upgrades ：
l MediaUni multi-integrated streaming media transmission network is upgraded based on the global real-time transmission network GRTN. It supports standard and low-latency live broadcast, real-time audio and video with one network. Diverse services such as calls, cloud rendering, data transmission, and remote control realize computing network integration and achieve a dynamic balance of high reliability and low cost.
l MediaCodec intelligent coding deeply combines traditional coding with AI to achieve the best balance of video quality, cost and computing power requirements from a business perspective, especially with the help of new generation coding Ali266's outstanding capabilities have achieved an average bandwidth saving of more than 48%.
l MediaService intelligent media service integrates media and AI capabilities at the frame level, completes the top-level design and evolution from traditional media processing to AIGC, uses AI to reshape business capabilities, and realize Intelligent, diversified and efficient media services.
l MediaBox terminal integrated suite meets the needs of rapid launch of audio and video capabilities in industry-specific and scenario-based situations. At the same time, new scenario AI capabilities are added to continuously implement micro-short dramas, New solutions for remote exams, corporate live broadcasts and other scenarios.
Media processing and service capabilities are the core of the video cloud, and are also areas where intelligence can be more fully penetrated. The interplay of the easy-to-use MediaBox terminal and the powerful MediaUni base is exactly the low-threshold realization and cost reduction of audio and video digital capabilities in thousands of industries. Effective and high reliability guarantee.
For the video cloud, no matter where AI intelligence is integrated in the entire link, whether it is a large model or AIGC, there is still a long way to go. The video cloud service may be an integrated platform tool, and where it is used is more important.
04 Implicit growth, explicit energy
After all, growth is the main theme of market expectations.
As for the market size of China's video cloud, IDC predicts that will reach US$19.16 billion in 2027.
We have observed that due to various comprehensive reasons, IDC has lowered the average annual compound growth rate of China's video cloud market. It is expected that the average annual compound growth rate from 2022 to 2027 will be 13.7%, but analysts also remind "The 2024-2026 growth forecast is still likely to be significantly revised. Changes in the traffic pattern of upstream audio and video platforms and changes in the way end users watch videos may have an impact on Forward market growth has had a significant impact."
At the same time, from a solution perspective, IDC believes thatintelligent demand may explode in the mid-to-long term, which is different from video production creation and media production. It is a stable growth model of asset management. It is assumed that "Audio and Video + AI" can complete a closed business loop on the industry application demand side and the technical service supply side in the medium and long term. Then "other" tracks may usher in rapid growth.
In fact, the real AI era has just begun. According to another IDC forecast, the global generative AI computing market will grow from US$820 million in 2022 to US$10.99 billion in 2026.
At the same time, among the top ten strategic technology trends released by Gartner in 2024, it was found that a common theme behind them is AI. Among them, Gartner predicts the two major trends of "Democratized Generative AI" and "AI-Augmented Development": by 2026, more than 80% of enterprises will use APIs or models for generative AI or applications that support generative AI are deployed in production environments, compared with less than 5% at the beginning of 2023; by 2028, 75% of enterprise software engineers will use AI coding assistants, At the beginning of 2023, this proportion was less than 10%. The growth momentum is astonishing.
Generative AI and AI enhancement technology are naturally inseparable from the current and future video cloud. Especially compared with image and text generation, the technical threshold of video generation is obviously higher, and the future space is also larger.
In the past, in the field of audio and video, we believed that the quality of AI output was far inferior to manual work, but this situation has begun to change subversively, whether it is the image quality restored by AI, the quality of materials generated by AI, or the quality of AI-generated materials. Media content can be understood like humans, and AI is even more sophisticated than humans in analyzing and refining video structure.Nowadays, it seems that all audio and video businesses can be redone with AI It's time.
Coincidentally, Alibaba Cloud Video Cloud grows on the two-wheel drive of Alibaba Cloud's "AI + cloud computing". Benefiting from this "the most open cloud in the AI era", the "Cloud Intelligence Reborn" video cloud has the potential to evolve for thousands of industries. possible.
05 Making Cloud Imagine
Finally, I would like to quote a passage from IDC analyst Wei Yunfeng to express his thoughts on the extension of the video cloud.
"Facing the future, it is difficult to predict in which year and in which scenario the demand brought about by 'imagination' will explode again, but higher-definition, more immersive, and more interactive terminal demands have always represented the direction pursued by end users in the hyper-video era. ; At the same time, widely distributed and connected video cloud edge resources and overall architecture, after the wave of intelligence sweeps across the world, may also provide more value under various grand and huge ideas, and even become the basis for the implementation of intelligent applications and services. catalyst."
Since IDC began to release authoritative data in 2018, Alibaba Cloud has maintained its No. 1 position in the overall video cloud market in China. Looking back on 2021, Alibaba Cloud Video Cloud staged an immersive summit with the theme of "Imagine", focusing on outlining the imagination of hyper-video. From then on and even earlier, it has been unwavering in its spatial imagination of "video + cloud". From now on, we will spare no effort to explore the new realm of "video cloud + AI".
If mathematics is the forefront of generative AI deduction, then imagination is the rudder of the video cloud.
In short, the AI era of video cloud has arrived.
