singlechip software development voice module

Introduction of intelligent sleep speaker

Smart sleep speaker is a product of speaker upgrade. It is a tool for home consumers to surf the Internet with voice, such as on-demand songs, online shopping, or weather forecast. It can also control smart home devices, such as opening curtains, setting refrigerator temperature, and heating the water heater in advance.

Control mode of intelligent sleep speaker

Non intelligent sleep speakers are generally controlled by remote control and body buttons. As for the body button, at least it has been replaced by a better "touch screen", and the remote control used by the current smart sleep speakers is increasingly regarded as a backup function. The current trend is to use mobile terminal applications as a control point, which is controlled by app/ WeChat / Alipay applet. In addition, intelligence will also be compatible with gesture recognition, voice control, wave sensing, etc. in general, the core goal of the control mode should be to simply and effectively transfer user commands to enhance the experience.

Communication means of intelligent sleep speaker

Intelligent audio transmits information through Bluetooth, Wi Fi, NFC and other information transmission methods. NFC is usually connected at the moment when the phone and speaker are connected. It is simple and direct, and there is no complicated use process. Similar to this, some intelligent sleep speakers also use the way of acoustic connection. Within a certain range of information transmission, quality and degree, Bluetooth and Wi Fi are equally matched, but the cost of Wi Fi is slightly higher, but for enterprises with intelligent front-end development, the difference can be smoothed.

Function of intelligent sleep speaker

Intelligent sleep speaker integrates some value-added functions. Integrated radio, weather broadcast, alarm clock and other functions are common. It integrates intelligent reading, intelligent land phone, even intelligent TV and intelligent voice robot assistant to carry out voice shopping, voice information query, voice video chat, etc.

How to solve several key technologies of intelligent sleep speaker

Speaker is the field of emphasizing brand and high technical threshold. The intelligent sleep speaker integrates acoustic design, wireless technology, speech recognition, far-field pickup, semantic analysis and many other technologies together. It is not only more complex in technology, but also more dependent on the support of music content platform. These factors are the problems that entrepreneurs need to face and solve. Of course, technology is still the fundamental. This paper hopes to analyze several key technologies and some unavoidable user experience problems from the existing popular products on the market, and give some references to entrepreneurs who are starting or preparing to enter this field.

(1) Small portable and bass enhancement technology

There are two main methods to enhance the bass of small box: one is the structural design of the box, such as sealed type, inverted phase type, labyrinth type, acoustic tube type and multi cavity resonant type. These structures need professional calculation and determination. Appropriate design can effectively improve the sound quality and bass effect of the speaker.

(2) Wireless technology and voice to code technology

Acoustics is always so wonderful, so acoustic researchers have long found a solution: acoustic communication code matching technology. This technology uses sound wave modulation technology to send the information needed for WiFi connection to the intelligent sleep speaker through the speaker of the mobile phone. The microphone configured by the smart sleep speaker receives the sound signal to demodulate and obtain the information, thus completing the configuration networking. The user only needs to input information on the mobile phone screen, which successfully solves the problem of the lack of screen in the smart sleep speaker Display and control issues. Sound to code technology is not too difficult in practice, but it needs a long time to accumulate to be stable and reliable.

(3) Far field voice wake-up and recognition technology

How can we solve this problem? Of course, we also have active and passive methods. Now let's talk about the passive method, which is the microphone array technology that we often talk about. Let's just talk about the technical difficulties of microphone array. Of course, many students will first think about the algorithm. It is a technical difficulty for multiple microphones to work together. In addition, the structure design and device are also the important factors that restrict the application of microphone array. The reason why this technology can not be applied until now is that it has been widely used. Of course, multi-channel acquisition technology is also a very important basic technology.

Think about it carefully. In fact, users' requirements for intelligent sleep speakers are far from simple as speech recognition can do. Obviously, it needs in-depth language interaction, which should also be the scene in your daily life. If language interaction can be achieved, then it can be said that intelligent sleep speakers can become the top priority in the whole smart home life scene. I believe that with the development of the times, that day will come.