In fact, the functions brought by voice AI technology are no longer limited to simple virtual assistants, but have developed more practical applications such as voice input and voice translation, which have penetrated into all aspects of our lives.
When Siri, Apple's voice assistant, was first launched, it caused quite a stir. At that time, even some private developers made corresponding Cydia plug-ins, so that iOS devices without official support could use the voice assistant function after jailbreak.
Voice assistant realizes human-computer interaction through voice and has the function of virtual assistant. This form was very novel at that time, which suddenly aroused the curiosity of many users.
Apple has made a start. Driven by strong market demand, other technology manufacturers have also launched voice assistant functions. 20 1 1 Google added voice search function to the search engine of Chrome browser to meet the voice input needs of users.
In 20 13, Cortana, Microsoft's voice assistant, went online on Windows Phone. Cortana shows the technical advantages that Microsoft has accumulated in the field of speech recognition for many years. As far as the experience is concerned, the voice effect of "Xiao Na" is very close to the real person, and it can almost achieve the effect of confusing the fake with the real, and it has countless fans at once.
At the same time, domestic voice AI companies are also actively launching similar products. For example, Iflytek launched the Voice Assistant App, which is also a partner of domestic mobile phone manufacturers such as Meizu and OPPO. Many voice assistants or smart assistants on users' mobile phones are using Iflytek's voice recognition technology.
However, when the voice assistant is rapidly popularized and matured on mobile phones, everyone's enthusiasm for it seems to be slowly fading. As far as my personal experience is concerned, it is rare to see people using the voice assistant function of mobile phones in public, although more and more manufacturers are greatly improving the priority of voice assistants, such as adding independent AI entity buttons and putting the switch of AI voice assistants in a conspicuous position on the desktop.
Many people, including myself, don't like to use voice assistants, because they are either not easy to use or inconvenient to use. At present, the voice recognition ability of mainstream voice assistants is generally strong, but there will still be a rollover phenomenon, especially in the case of noisy environment, vague voice semantics (or inaccurate pronunciation of Mandarin). In many cases, direct manual operation will be much more convenient than calling out a voice assistant to help.
It is easy to understand that it is inconvenient to use. Using voice assistant in public places, on the one hand, we should overcome the shame of talking to mobile phones in public places, on the other hand, we should take care of our privacy not to be exposed.
At present, smart phones are quite popular, and our proficiency in the operation of this necessity is already high. Many times, we don't need the interactive form of voice to realize various functions. In this case, the voice function on the mobile phone sometimes does give people a feeling of chicken ribs.
Voice function shines brilliantly in the Internet of Things
Voice AI technology does not stop at smart phones, but extends to other products. Nowadays, whether it is a smart speaker, a smart TV or an endless stream of intelligent hardware products, as long as it is labeled as "smart", the function of voice control is essential.
Interestingly, in the home environment, the advantages of voice AI technology seem to be more fully released. In more private scenes, our willingness to use voice interaction is greatly enhanced. In a relatively closed environment, users don't have to worry about privacy, so their psychological preparedness will be reduced, and the possibility of trying voice interaction will naturally increase.
For products such as TV speakers, the traditional control tools are nothing more than physical buttons or remote controllers. In contrast, the advantage of voice lies in liberating users' hands. In the home environment, users can directly use voice commands to realize audio and video playback, home control and other functions in the scene where they can't do housework, which undoubtedly greatly improves the efficiency and experience.
In recent years, both smart screens and some brands of high-end TVs have enhanced the pickup effect of large-screen equipment. By adding a radio microphone, users can wake up TV equipment as an intelligent control center anytime and anywhere. At present, the industry generally believes that the popularity of 5G and broadband in the future and the further development of AI technology will bring us into an era of Internet of Everything.
When all the electrical equipment we can touch can be networked and have intelligent functions, how to control them conveniently will become the most critical issue. At present, pronunciation is the lowest learning cost and the most convenient way to use.
5G, AI, IoT, internet of everything ... about the future, ambitious technology manufacturers don't want to miss it. Apple, Google, Xiaomi and other companies. They are constantly strengthening the presence of voice AI technology in mobile phone systems, largely by laying out in advance, constantly cultivating users' habits, and allowing target groups to unconsciously integrate into the ecosystem established by manufacturers.
For a specific group of people, pronunciation is a revolutionary technology.
In addition, it is undeniable that for some specific people, voice AI related technologies have an important role in promoting the mobile Internet and smart life.
For many "elderly" users, typing with virtual keyboard on mobile phone is too expensive and difficult to learn. It is precisely because of this that many elders like to send long and long voice messages when chatting on WeChat, which annoys many people and makes them want WeChat to turn off the voice function.
But now social applications such as WeChat, as well as some third-party input methods, already support voice input function, which can turn voice into text. Judging from the current technical level, the recognition accuracy and availability of mainstream voice input have been quite high. For small screen devices such as smart watches, voice control is often much more convenient than touch operation.
In addition, for some visually impaired people, the development of voice AI technology has greatly lowered the threshold for them to use smart devices and enjoy mobile Internet life. We have tested some mainstream applications in daily life before. Through the interaction of voice and physical keys, the blind can also easily complete many operations and use many functions.
Whether you like it or not, the development of internet technology in recent years is a process of constantly encroaching on users' privacy space. Even Apple, which has always advertised respect for users' privacy, was exposed to privacy issues on 20 19. User voice data uploaded by Siri can be obtained and analyzed manually.
The balance between privacy and technology is also a problem faced by companies developing voice AI. Theoretically, in order to make the voice function easier to use and more intelligent, it is necessary to analyze and optimize the voice data of users. The key question is how likely the collected information is to be leaked and illegally used.
Last year, Accenture, a consulting firm, conducted a survey of users in China. The data shows that China users' satisfaction with voice assistants is as high as 97%, but their trust is still not high. China users' main concerns about voice assistants are security issues, users' needs not being understood and privacy issues.
Moreover, after some negative things, such as the theft of webcams and the sale of room opening data, domestic consumers are not low on privacy issues.
In 20 18, the EU passed the most stringent personal data protection regulation GDPR, which made it clear that users have absolute control over personal data, and the penalties for corporate violations are extremely heavy, with a minimum fine of10 billion euros.
From the user's point of view, this may be a good thing, which can curb the abuse of user data and invasion of privacy by enterprises. However, too strict restrictions will also make it more difficult for technology companies to promote technological progress that requires big data, such as voice AI.
If we look further, we believe that in the future technological life, the importance of voice in human-computer interaction will be greatly enhanced, and it will even become the most important operation mode in smart home and other scenes. However, in this process, we hope that our personal data can also be handled better.