Along with the rapid development of the Internet of Vehicles, in order to further enhance the voice interaction experience of smart cars, Jietong Huasheng has launched an enhanced Lingyun speech recognition technology. The technology adopts the latest deep-learning algorithm for anti-noise developed by Jietong Huasheng. By modeling the driving environment noise information, it effectively processes the driving environment noise such as tire tire noise, wind noise, car air conditioning sound, engine sound, etc. The algorithm is further trained and optimized by using a large amount of actual driving noise data, and has strong anti-noise ability, and can have a very high speech recognition rate even in the case of a vehicle speed of 80 km/h.
Jietong Huasheng said that the enhanced Lingyun speech recognition technology has been integrated into the Lingyun far field voice interaction solution. By combining with the multi-microphone array, it can “filter†most of the vehicle environment noise and realize long-distance voice. Identification, after practice testing, its extremely high recognition rate fully meets the requirements of the driving environment for far-field speech recognition, and realizes smooth and natural voice control of intelligent on-board equipment. At the same time, Jietong Huasheng cooperates with domestic intelligent vehicle equipment manufacturers, solution providers, software developers and other industry partners, adheres to the development concept of “cooperation and win-winâ€, and provides microphone array and cloud+end voice interaction for partners. The in-vehicle voice solution integrating software and hardware, such as content service, jointly promotes the popularization of intelligent voice interaction technology in the automotive field!
Enhanced Lingyun speech recognition + microphone array easily handles driving environment noise
For a long time, the voice interaction technology is affected by multiple complex factors such as background noise, echo, reverberation, etc. The driver must speak at a close distance to the microphone “bite the earâ€, and the recognition rate is low, which can not meet the recognition of the vehicle equipment on the road. The requirements of the driver's order severely restrict the application of voice interaction technology in the field of intelligent vehicles.
The Jietong Huasheng Lingyun far field voice interaction solution can solve this problem well. The solution includes multi-microphone array and enhanced speech recognition technology. The multi-microphone array accurately picks up the speaker's voice signal, suppresses noise and reverberation, and enhances the speech recognition technology to further process the residual car noise, greatly improving the speech recognition rate in the driving noise environment.
The multi-microphone array can accurately calculate the angle and distance between the sound source and the array module according to the time difference of the sound arrival, realize the positioning and tracking of the target sound source, and utilize the spatial filtering characteristics of the microphone array to form in the target speaker direction. The pickup beam picks up only the beam signal, suppresses noise and reverberation (reflected sound) outside the beam, and at the same time, filters out the sound emitted by the audio device through the echo cancellation technology, so that the voice recognition engine obtains a clearer user voice. Pure.
Since the microphone array will inevitably pick up part of the driving environment noise and affect the speech recognition rate, Jietong Huasheng comprehensively considers the physical structure and vehicle speed changes in the car, the car tire noise, wind noise, and car air conditioning sound during driving. Vehicle noise information such as engine sound is modeled, and a large amount of actual driving noise data is used for training. The speech recognition engine is further optimized, and an enhanced Lingyun speech recognition technology for intelligent vehicle is introduced to easily cope with common vehicle environment noise. The technology adopts the deep neural network algorithm (DNN) algorithm, which can continuously improve the speech recognition rate by using the accumulated speech data to train and optimize the algorithm during the user's use. Through this series of technologies, the speech recognition rate in the vehicle noise environment is greatly improved, and the requirements of human-computer interaction for speech recognition technology in the actual driving process are fully satisfied.
Enhanced speech recognition enables voice interaction to truly meet driver needs
Lingyun far-field voice interaction solution that combines enhanced Lingyun speech recognition technology provides partners with software and hardware integrated car audio solutions such as microphone array, cloud + end voice interaction and content service. The solution provides a unified interface to quickly establish the connection between voice commands and in-vehicle systems and various applications. With open service extension and flexible capability support, the smart car has voice wake, sound source orientation and far field noise reduction. , echo cancellation, voice interruption, voice correction, multi-round dialogue and other functions, to create a smart car device that can listen, speak, and think, completely break the industrial bottleneck of the application of voice interaction technology in vehicle equipment, and truly realize Natural, smooth human-machine voice interaction.
During the driving process, when the driver is disturbed by traffic jams, waiting, red lights, etc., as long as the music and songs you want to listen to are spoken, the car equipment will automatically retrieve and play the music, alleviating the anxiety of the driver waiting; Come in, and when the driver is driving, just say "answer", you can connect to the phone, truly realize that "the gentleman does not move", no longer need to "press, find, find" in the process of driving Return the driver's hands and eyes to the driver to ensure safe driving.
Lingyun far field voice interaction solution has fully supported calling, sending and receiving text messages, listening to music, listening to radio, map navigation, POI, vehicle condition inquiry, weather inquiry, service inquiry, equipment control, traffic system question and answer, etc. in the field of intelligent vehicle. It provides users with free access to more than 30 kinds of information and content services commonly used in daily life such as weather, transportation, news, stories, stocks, music, trains, flights, restaurants, hotels, etc., to fully meet the various needs of drivers in the car scene. . At the same time, Lingyun far-field voice interaction solution supports interrupting at any time, only listening to the information that the driver really wants to express, and through voice error correction and multi-round dialogue, the voice interaction is smoother and more natural, and truly meets the driver's needs.
Jietong Huasheng insists on cooperation and win-win and shares opportunities with partners
The enhanced Lingyun speech recognition technology is derived from the first all-round artificial intelligence open platform launched by Jietong Huasheng in 2011, the cloud cloud platform (hcicloud.com). After four years of development, the Lingyun platform has become a domestic labor supply. The artificial intelligence open platform with the most intelligent capabilities and the widest service users. Since its launch, Lingyun platform has been fully open to the industry. All partners, including individual developers, can acquire technical capabilities such as intelligent voice, intelligent image, intelligent semantics and biometrics through the Lingyun platform.
With the rapid development of IOT, in 2014, it is more convenient for domestic smart car, smart device, home appliance, robot and other terminal equipment manufacturers, integrators and developers to use the full range of artificial intelligence capabilities of the cloud cloud platform (hcicloud.com). Jietong Huasheng has launched the Lingyun intelligent terminal interaction solution for enterprise-level developers: Lingyun Seed. Lingyun Seed fully integrates many artificial intelligence technologies of Lingyun platform, including Lingyun speech recognition, speech synthesis, semantic understanding, voiceprint recognition, face recognition, fingerprint recognition, optical character recognition, etc., and provides developers with Easy to use development interface. By applying Lingyun seeds and combining multi-microphone array modules, developers can easily implement intelligent human voice interaction, voice wake-up, voice navigation, text and license image recognition, voiceprint unlocking, face recognition and other human-computer interaction functions. Multi-angle to meet the market demand for diversified artificial intelligence technology capabilities of terminal equipment such as vehicles, home appliances, home appliances and robots.
A flower opening is not a spring, Jietong Huasheng will adhere to the development concept of "cooperation and win-win", share the opportunities of artificial intelligence industry with the vast number of partners, and jointly create a good artificial intelligence industry ecology. In order to better serve the nation's smart terminal equipment partners, Jietong Huasheng will further accelerate the process of establishing branches and offices throughout the country. For example, in South China, smart car, smart home, robot development and manufacturing companies gather and own Unique advantages such as strong innovation and high technology. In order to do a good job of technical support for partners, in 2015, Jietong Huasheng has set up branches in Shenzhen and Xiamen to provide more timely and better technical support for partners.
In the future development, Jietong Huasheng will continue to rely on the research strength of artificial intelligence of Tsinghua University, and focus more on the promotion and application of artificial intelligence technology research and industrialization such as intelligent voice, and jointly promote the Chinese intelligent terminal through the sincere cooperation with partners in the industry. The equipment fully serves the public and accelerates the pace of the intelligent life era!
It is the thinest phone case made from bullet-proof materials. It is the world's first modular magnetic wallet made from carbon fiber.
It is all about your carrying experience. It is a magic world of seamless carrying experience. And that makes you being different from anyone else.
Exclusive Material. Slim Fit and Sensuous With "Body Armor Grade" material, made of 100% Aramid Fiber. Now, your phone is protected with the same material used in body armor, spacecraft, jet engines, and Formula 1 Race cars. Yet, it retains the thinness of your iPhone the day you purchased it -- before it was covered up with an thick bulky case.
Wireless charging friendly. Bring you the best carrying experience and wireless charging experience by using with PITAKA magnetic car mount(Magmount Series), the unique transit free design with metal plates inside makes you can use PITAKA car mount freely without sticking adhasive metal plate on your device. Yet,no interference to your wireless charging,WIFI,GPS,Apple Pay or your signals.
A Soft 3D-Grip touch. Texture you need to feel to believe. It almost clings to your hand without feeling sticky. A protective three-layer coating and manual polishing is applied to each phone case.
Enjoy your "almost naked" iPhone with a case that is thinner than a dime. Thickness 0.03in. Min. Weight 0.49oz. So light that you even will not feel the weight increase to your phone. Well protect phones from most of the daily drops.
iPhone Case,Waterproof iPhone Case,iPhone 7 Case,iPhone 8 Case
Shenzhen Aokal Technology Co., Ltd. , https://www.aokals.com