4+1 ring five microphone array creates a new pattern of voice interaction

The origin of artificial intelligence was at the Dart Month conference in 1956. It was an artificial program that could be perceived, hostage, decision-making, and executed like human beings. Nowadays, artificial intelligence is becoming the trend of the times, and gradually enters the eve of the real explosion. At present, AI has caught up with humans in computing intelligence and perceptual intelligence. Alphago is a good example of victory over Li Shishi, and cognitive intelligence is the key challenge of current artificial intelligence.

To this end, on March 30th, HKUST News held a seminar on “Xinfei Core Technology Development Day – Maiyun Cloud, Leading Smart Interactive Newism” at the Akime Internet Community in Shenzhen, sharing the understanding of artificial intelligence and This effort has resulted in a complete intelligent hardware voice interaction solution and an open platform core interaction technology.

Photo: Ma Hanjun, deputy general manager of Keda Xunfei Open Platform

Ma Hanjun, deputy general manager of Keda Xunfei Open Platform, said that artificial intelligence is divided into three stages, namely computing intelligence, perceived intelligence and cognitive intelligence. At present, computational intelligence has surpassed a human model. The calculation of the atomic bomb from the beginning has already represented the ability of computing; the perception of intelligence, the robot is approaching humans quickly, and the United States is also testing robot dogs and Google unmanned vehicles. Cognitive intelligence, this is the current challenge of artificial intelligence, that is, how to make the machine learn and understand knowledge.

At the same time, it was pointed out that in the last year, Xunfei released AIUI, which is an intelligent interactive solution tailored for intelligent hardware. It redefines the human-computer voice interaction standard in the era of Internet of Everything, with far-field noise reduction, dialect recognition, and speech correction. , multi-round dialogue and other functions, through a unified interface, open service expansion, flexible capabilities to achieve a combination of human-computer interaction and product experience. HKUST News hopes to work with partners to change the world with artificial intelligence.

Photo: Zhang Liangchun, Product Manager, Feiyun Platform Business Unit, Keda News

In the field of artificial intelligence, the University of Science and Technology News has brought the latest weapon to break the bottleneck of far-field voice interaction - the ring five microphone array. Zhang Liangchun, product manager of the company's Feiyun platform business unit, said that the five-microphone array is a four-plus-one ring. It is a very good requirement for vocal recognition in this range. 5 meters, support 360 full-plane pickup angle, support continuous wake-up, echo cancellation, voice interruption.

4+1 ring five microphone array product mode

Zhang Liangchun pointed out that using the microphone array module of Keda Xunfei can help developers to develop products quickly, greatly saving product development cycle, and also provide soft core solutions including single wheat, double wheat, four wheat linear, ring five wheat and ring seven wheat. In particular, the ring seven wheat program supports 7 meters pickup distance, 360° sound source positioning, ±10° sound source positioning accuracy, and the audio signal-to-noise ratio of the array input is higher. At the same time, the hardware solution with the voice wake-up closed-loop optimization service, speech recognition, semantic understanding deep customization, identity authentication, natural speech synthesis and other functions of the Keda Xunfei open platform can make the developer's products as bright and outstanding as the fireflies in the dark.

This is a product model for the five-microphone array, which is mainly to change the voice data and audio processing on a single hardware. There are five microphones that simultaneously record 5 audios, which are sent to the hardware module to process the information, and the machine needs to supply power to it. Zhang Liangchun said that the advantage of this module is that it can be quickly integrated, can quickly form a product, has a fast word, has advantages under normal conditions, and has some shortcomings. The downside of it is that the cost is a bit high and only suitable for small batch development.

In the five-microphone array soft core scheme, there are two aspects worthy of attention. One is audio collection, from audio collection to audio aggregation to processing, each node will provide products to help users to better integrate software; the other is audio aggregation, basically audio processing CPO or IPO Processing audio can only handle mobile clouds, and it also requires an audio aggregation circuit for processing.

Fiber optic adapters are typically used to connect two fiber optic cables together. They come in versions to connect single fibers together (simplex), two fibers together (duplex), or sometimes four fibers together (quad). Adapters are designed for multimode or single-mode cables. The single-mode adapters provide more precise alignment of the tips of the connectors and can be used to connect multimode cables. The reverse is not suggested as this might cause misalignment of the small single-mode fibers and loss of signal strength (attenuation).

Optical attenuators are commonly used in fiber-optic communications, either to test power level margins by temporarily adding a calibrated amount of signal loss, or installed permanently to properly match transmitter and receiver levels. Sharp bends stress optic fibers and can cause losses. If a received signal is too strong a temporary fix is to wrap the cable around a pencil until the desired level of attenuation is achieved. However, such arrangements are unreliable, since the stressed fiber tends to break over time.

Types:

Optical attenuators can take a number of different forms and are typically classified as fixed or variable attenuators. What's more, they can be classified as LC, SC, ST, FC, MU, etc. according to the different types of connectors.

Sijee provides different types of Optical Fiber Attenuator, Fiber Optic Attenuator, Fiber Optic Fixed Attenuator, FC Fiber Attenuator.

Fiber Optic Attenuator

Optical Fiber Attenuator,Fiber Optic Attenuator,Fiber Optic Fixed Attenuator,FC Fiber Attenuator

Sijee Optical Communication Technology Co.,Ltd , https://www.sijee-optical.com