Artificial intelligence voice assistants are being integrated into cars to achieve free and intuitive functions. Voice assistants such as Apple Siri can recognize and respond to natural language orders, so that drivers can interact more effectively with vehicles. It is complicated to integrate natural voice virtual assistants and require a lot of professional knowledge in resources and learning and data collection. Therefore, only a few companies can successfully achieve this.
The popularity of ChatGPT has prompted many people to think about AI's potential applications. One of them is in the automotive field. With the simplification of the vehicle dashboard, the trend of integrating more functions to the central display, such as entertainment, climate control and vehicle diagnosis. The central computer in the vehicle is becoming more and more powerful, and can do more. All of this allows the driver to interact with their vehicles in a simpler and more user -friendly way, and at the same time enable more advanced and customized functions for the vehicle itself.
In addition, this matches the development of software definition vehicles, the latter controls all vehicle functions by using a centralized software architecture to make this integration further. This allows greater flexibility and the ability to update the vehicle system through the air (OTA).
People are increasingly needed to integrate additional functions into the central display, such as voice assistant, car digital assistant, and other senior driver assistance systems (ADAS). However, excessive simplification can lead to many problems. Although modern cars generally use touch screen displays, some people still like to use knobs or buttons in the car cab. The following are some reasons:
Touch Feedback: Many people find that using these physical controls is more intuitive than browsing the digital menu on the touch screen display. The knobs and buttons will provide physical feedback when they are pressed or rotated, which allows you to interact easily with the control without the need to remove the sight from the road.
Visible: In some cases, the knobs and buttons are more likely to see and use under the bright sun or other challenging light conditions, because they are not affected by glare or reflection like the touch screen display.
Safety: The use of physical knobs and buttons is safer to interact with the touch screen display because it allows the driver to put their hands on the steering wheel and stare at the road.
Therefore, installing a user -friendly, reliable and intuitive simplification of the human -machine interface (HMI) on the central screen of the car is critical to minimize the driver's learning curve and enable them to easily and efficiently access the required need Functions without any errors. The most important one is the virtual voice assistant.
There are several popular virtual voice assistants in the market today, such as Amazon Alexa, Google Assistant, Apple Siri, Microsoft Xiaona, Samsung Bixby, Baidu Duer and Xiaomi Xiaoli. In addition, there are other proprietary virtual voice assistants designed for the automotive industry, such as CERENCE, Soundhound house, Harman Ignite and Nuance Dragon Drive.
Most of the virtual assistants in the automotive industry are integrated with the vehicle information entertainment system, providing drivers with various voice control functions, including recreational telephone, weather forecast, music flow and voice control navigation. In addition, they aims to identify and respond to natural language orders so that drivers can interact with vehicles in a more intuitive and easier way. By providing a safe and convenient way to interact with the vehicle, these virtual voice assistants allow the driver to put their hands on the steering wheel and stare at the road.
Although the virtual voice assistant has improved significantly in recent years, there are still some challenges to be resolved. The following are some common problems currently existing virtual voice assistants:
Understanding complex commands: Virtual voice assistants may encounter difficulties when understanding complex commands or requests involving multiple variables or conditions.
Voice and dialect: Virtual voice assistants may also be difficult to understand users with different accents or dialects.
Background noise: virtual voice assistant may be very sensitive to background noise, which makes it difficult for them to understand the user's command or request.
Privacy issues: As virtual voice assistants become more and more common, people are more and more worried about the privacy of user data.
Integration with other automotive system: virtual voice assistant may be difficult to integrate with other systems or equipment, which may limit its function and practicality.
ChatGPT can be said to be natural language and talks like humans, because it is a language model that uses deep learning technology called the Transformer architecture to train a lot of text data. During the training, ChatGPT came into contact with a large amount of natural language text data, such as books, articles and web pages. This enables it to learn the patterns and structures of human language, including grammar, vocabulary, syntax and context.
Different from the widely foundation training methods, natural language training (such as training provided by ChatGPT) allows the development of models that fine -tuning for specialized datasets, which may include commonly used vehicle commands or a series of different ethnic accents. Then fine -tune the model by further training models on the large -scale unsigned data corpus to improve its language understanding ability.
In general, the potential of car natural language voice dialogue assistants is huge. With continuous research and development, we can expect to see more advanced and more complicated voice assistants in the future. It is a complex and time -consuming process to develop a successful use of a natural language virtual voice assistant for cars. It requires multiple iteration training and fine -tuning.
Because development requires a lot of data, computing resources and professional knowledge, only a few companies such as Microsoft, Tesla, Nvidia, Qualcomm, Google and Baidu have resources to take on this work. The development of this technology is estimated to take three to four years. The demand for vehicles above level 3 will increase.
Just as our report, "Should the original equipment manufacturer of automobiles enter the production of autonomous driving chips?" The automotive industry will face obstacles related to electrification and intelligent technology, and requires continuous capital investment and semiconductor suppliers. Therefore, only a few veteran car manufacturers with considerable economies can provide funds for these measures. The increasing popularity of natural voice control in cars will only exacerbate these challenges.
Reposted from: International Electronic Commerce, automatically translated by Google
