Market and Technology Trends
Computing and AI Technologies for Mobile and Consumer Applications 2021
By Yole Intelligence —
Penetrating everyday products will see the market for AI technologies for the consumer market reach $5.6B in 2026.
What's new:
- Inclusion of wearables markets, including smartwatches, True Wireless Stereo (TWS) earbuds, audio headsets, Augmented Reality (AR) and Virtual Reality (VR) headsets with market trends, ecosystems, forecasts, and technology trends
- Hardware technology trends analyzed in detail
- Wider scope including most computing processors, not only focusing on AI. Forecasts for the main processors for all applications studied in volumes shipped, Average Selling Price (ASP) and revenue
Glossaries 2
Definitions 3
Table of contents 7
Report objectives 10
Scope of the report 11
Report methodology 14
About the authors 15
Companies cited in this report 16
What we got right, what we got wrong 17
Who should be interested by this report 18
Yole Group of Companies related reports 19
Three-page summary 21
Executive summary 25
Context 71
- Consumer electronics overview – on the road to augmented human
- From image processing to image analysis
- From audio processing to audio computing
- Toward edge computing
- Artificial intelligence and beyond
Market forecasts 102
- Market forecast scope
- Market forecast methodology
- SoC volume shipment forecast 2016 – 2026
- SoC ASP forecast 2016 – 2026
- Mobile and consumer SoC revenue forecast 2016 – 2026
- AI penetration rate
- Mobile and consumer SoC with AI unit shipment forecast 2016 – 2026
- AI unit ASP forecast 2016 – 2026
- Mobile and consumer integrated AI unit revenue forecast 2016 – 2026
- Mobile and consumer computing and AI forecast summary
Market trends 118
- Smartphone
- Mobile device history
- Mobile market forecast 2016 – 2026 in units
- Evolution of smartphones towards AI
- Why use artificial intelligence in smartphones?
- Augmented reality in smartphones
- Virtual personal assistant in smartphones
- Smartphone application processors
- What about audio hardware for AI in smartphone?
- Drones
- Introduction to the consumer drone system
- Consumer drone market forecast in units
- Artificial intelligence in drones
- The difficult requirements for autonomy
- Are drones still an interesting business?
- Smart Home
- Smart home definitions
- Where can you find artificial intelligence in a smart home?
- Smart assistant trends
- Smart TV trends
- Consumer security camera market
- Consumer security camera market drivers
- Emerging smart home market trends
- Wearables
- Wearable market trends – the three waves
- Smartwatches
- Smartwatch market forecast 2016 -2026 in units
- Smartwatch market trend
- Smartwatch main use cases
- True Wireless Stereo (TWS) Earbuds
- TWS earbuds trends
- Definition and TWS earbuds market forecast 2016 -2026 in units
- Treatments trends to improve sound quality
- AR and VR
- AR and VR definition
- AR – What do consumers want?
- AR use cases for consumers
- AR market trends
- Is AR overhyped? Will AR happen for the consumer?
- VR use cases
- Market trends conclusion
Market shares and supply chain 191
- Companies offering an AI solution
- IP and design services companies for consumer market processors
- Design companies for consumer market processors
- Consumer AI noteworthy news of the last three years
- Consumer processor ecosystem overview
- Smartphone
- Smartphones OEM landscape
- Smartphone market share
- Smartphone application processor makers landscape
- Why would an OEM want to create its own chip?
- OEMs: Who can make their own APUs?
- Application processor revenue
- Smartphone processor supply chain 2019-2021
- Main player focus
- Drones
- DJI is the leader of consumer drone market
- Smart Home
- Smart speaker landscape
- Focus on top player strategy
- Smart speaker processor supply chain and landscape in 2020
- Smart TV landscape: OEM and processor designers
- Security cameras: OEM and processor designers
- Smart home player acquisitions and investments
- Wearables
- Smartwatch player landscape
- Smartwatch processor supply chain 2021
- Smartwatch application processor maker landscape
- TWS earbud players
- TWS processor supply chain 2021
- TWS Bluetooth processor landscape and strategy
- AR and VR wearable device landscape
- AR and VR processors
- Wearables player acquisition and investments
- Conclusion
- Tech giant analysis
- Company profile
Technology trends 278
- Why is computing AI at the edge a necessity?
- Smartphones
- AI accelerator, the specialized unit
- Main player descriptions
- High-end application processors – specifics
- Status of AI for imaging and audio in smartphones
- Drones
- Smart homes
- Smart speaker processor technology trend
- Smart speaker and security camera processor analysis
- High-end processors for smart homes – specifics
- Computing hardware for systems used in smart homes
- Wearables
- Smartwatches, TWS earbuds and AR/VR headset processor analysis
- TWS earbud processors specifics
- Technical trends for TWS Bluetooth processors
- Technology trend conclusions
- Hardware technology trends conclusions
Conclusion 221
Appendix 325
- Technology software
- Processor definition
How to use our data 351
About Yole Développement 352
AI FEATURES ARE BECOMING STANDARD IN CONSUMER APPLICATIONS
Artificial Intelligence (AI) technologies are already in many products that we use every day. They started in our smartphones, through applications such as face recognition or photography setting optimization. They are now spreading into most consumer applications, from smart home cameras with features such as intrusion detection to smart assistant support integrated into earbuds. Today imaging AI and audio AI are the most widespread applications. For several reasons including privacy protection and low latency requirements, AI inference is increasingly computed at the edge of the network, in the actual consumer device. This requires special processor capabilities to run these operations in real time. The share of processors dedicated to run AI is therefore growing. In this report, we analyze and forecast both the processor and AI unit markets, including the area of silicon specifically dedicated to accelerating AI operations. Figure 1 shows that the processor market has a 7.6% Compound Annual Growth Rate from 2020-2026 (CAGR20-26) driven by dynamic applications like earbuds. AI unit revenue has a 24.7% CAGR20-26, boosted by strong penetration of AI into most consumer applications.
AI UNITS FOLLOW THE INTEGRATION TREND IN THE CONSUMER MARKET
Processors for consumer markets are today mostly Systems on Chips (SoCs). They integrate all the required functions, such as Central Processing Unit (CPU) cores, audio Digital Signal Processing (DSP), and Bluetooth connectivity, rather than having a standalone chip for each function. Other options also exist and are analyzed in this report, but this is the main trend and it also affects AI hardware. Most consumer processor players are directly integrating AI units into their SoCs to accelerate AI inference. This sector is unlike CPUs and Graphics Processing Units (GPUs), where many solutions from Intellectual Property (IP) players such as ARM contribute. Instead, the processor designers are mostly not using solutions developed by IP companies, such as the new ARM Ethos solution. They prefer developing custom IP, as it is a good way to differentiate themselves from their competition.
BETWEEN SMARTPHONE OEM AND FABLESS PROCESSOR COMPANIES, FEW PLAYERS COMPETE ON HIGH-END TECHNOLOGIES
There are two kinds of processor players. One is the smartphone-making OEMs with processor design capabilities, such as Apple, Samsung, Huawei, and soon Google. The other is fabless processor companies such as Qualcomm, MediaTek, and UniSoC. Qualcomm and MediaTek are the two biggest fabless processor companies for consumer applications, both in terms of shipments and revenues. They have products targeting most consumer applications, with either one or the other usually being the leading player. Apple follows a different model, however. It is the third-placed processor maker for consumer markets, but doesn’t sell processors to anyone. Processors let Apple control its ecosystem and to adapt its products’ hardware to its software. Even if Apple’s silicon design activity is just to support its products, it designs toplevel processors and frequently drives innovation in this domain. Following a similar strategy, HiSilicon, a Huawei subsidiary, was very dynamic in 2020, following the strong growth of Huawei products. However, following the application of US sanctions, the story will be totally different. We expect HiSilicon to lose almost half of its Application Processor Unit (APU) revenue in 2021. In the short term, the future of high-end HiSilicon products made technology beyond the 14nm lithography node is uncertain. Smartphone processor designers also lead the race in most other consumer product processor markets. That is especially true in smartwatches which has the same players, both for end-products and processor. For True Wireless Stereo (TWS) earbuds, the story is different. Big processor players compete with historical processor players specialized in Bluetooth and audio technologies, but also with new players, mainly from China. For example, BES Technic has leveraged the skyrocketing growth of TWS earbuds to become a major player in this market. Smart speaker makers are also different from big smartphone players. They are US and Chinese tech giants, through partnerships with processor players. For example, Amazon and MediaTek make a processor that integrates Amazon’s custom AZ1 Neural Edge processor. Other players include Synaptics, Amlogic, and Allwinner Technologies. All the big players, and US and Chinese tech giants especially, are very involved in acquisitions of and investment into AI start-ups. This entire dynamic ecosystem is analyzed in detail in this report, including a broad analysis but also specific focuses on wearables and smart homes.
Airoha, Alibaba, Allwinner Technology, Alphabet, Amazon, Ambarella, Amlogic,AMD, Analog Devices, Anker, Apple, ARM, Asus, ATI, Axis Communication, Baidu, Bes Technic, Bluetrum, Bose, Broadcom, CEVA,Cirrus Logic, Cray, Cypress, Deephi Tech, DeepMind, DJI, DSP Group, Edifier, Facebook, Fossil, Fitbit, Garmin, Greenwaves, Google, Graphcore, Hailo, Hisilicon, Hover Camera,Honor, HTC, Huami, Huawei, IBM, Imagination, Infineon, Instagram, Intel, Inventiv, Jabra, Jaybird, Jieli Technology, Kealtek, Knowles, Lenovo, Lemfo, LG, LightOn, Magic Leap, Mediatek, Microsoft, Mobvoi, Motorola, Nokia, Nordic Semiconductor, Novatek, Nuance, Nubia, Nvidia, Nuvia, NXP, Oculus, Oppo, Omnivision, OnePlus, Pacific Future, Parrot, Perceive, Pico, Qualcomm, Realme, Rockchip, Samsung, Skydio, Silicon Labs, Socionext, Sonos, Sony, STMicroelectronics, Synaptics, Synopsis, Tencent, Texas Instruments, Toshiba, TSMC, UniSoc, VeriSilicon,Videantis, Vivo, Xiaomi, Xilinx, Zepp, ZTE, and many more.
Key Features:
- Types of hardware for smartphones, drones, smart home devices including smart speakers, smart TVs and cameras and wearable devices including smart watches, TWS earbuds, and AR/VR headsets
- Consumer processor forecast in volume shipments, ASP and revenue
- Forecast for AI penetration rate into main processors, with unit ASP and revenue
- AI technologies used in consumer applications
- Edge computing for AI
- Ecosystems, market forecast, and trends
- AI consumer technology trends, including hardware evolution and AI integration
- Investments, mergers and acquisition analysis
Objectives of the report:
Provide an understanding of computing trends and dynamics for key mobile and consumer applications:
- Market dynamics and segmentation breakdown by application and end-system
- Systems - volume shipment forecast
- Hardware for consumer application - revenue forecast, volume shipment forecast, ASP
- Type of hardware associated with each end-systems, key trends and dynamics
Provide a scenario for AI within the dynamics of the consumer market, and understand AI’s impact on the semiconductor industry:
- Hardware for AI - revenue forecast, volume shipment forecast
- Focus on consumer applications with embedded technologies: smartphones, consumer drones, smart speakers, smart TVs, home security cameras, smartwatches, TWS earbuds, AR and VR headsets
Deliver an in-depth understanding of the ecosystem and players:
- Who are the players? What are the relationships inside this ecosystem? What happened in the last three years?
- Who are the key suppliers to watch, and what technologies do they provide?
Offer key technical insight and analysis into future technology trends and challenges:
- Key technology choices
- Technology dynamics
- Emerging technologies and roadmaps