标签： 上海桑拿论坛

Author | Cheng Qian

Edit | Desert Shadow

Voice interaction has been played with new tricks? !

Wisdom reported on April 2nd that on Monday, Baidu released the industry’s first end-to-end speech language model based on new Cross-Attention, which has been launched online for free.

Summarize the newly upgraded essays with several key words: voice interaction.Ultra-realistic, ultra-low delay and ultra-low cost.

Super-fidelity refers to the fact that Wen Xiaoyan, equipped with a large phonetic language model, can understand Chongqing, Guangxi, Henan, Guangdong, Shandong and so onCharacteristic dialect can also realize emotional dialogue.; Ultra-low latency means that the waiting time of users can be reduced from the common 3-5 seconds in the industry toAbout 1 second, almost the same as talking to real people; Ultra-low cost means that in the voice question-and-answer scenario of telephone voice channel, the call cost is about lower than the industry average.50%-90%。

Jia Lei, chief architect of Baidu Voice, revealed that the model can be deployed on L20 card, and under the condition of meeting the delay requirement of voice interaction,The concurrency of dual L20 cards can be more than several hundred.. At present, the training process of speech language model is convenient, which can be realized by optimizing hundreds of cards in one week based on Wenxin model, and the optimization work itself is not complicated.

Compared with the application of big model in voice interaction scene, what is the uniqueness of this big model of voice language? How to reduce the call cost to 90% at the highest? How to interpret the innovation behind it? Zhi Dongxi had an in-depth exchange with Jia Lei, chief architect of Baidu Voice, trying to find the answers to these questions.

First, the real-life dialogue experience, the upgraded version of the text Xiaoyan voice interaction is more silky

The development of large-scale model in voice interaction scene is evolving towards a more natural, low-delay and highly realistic voice interaction experience. And this more anthropomorphic interactive experience, we get a glimpse of the embryonic form in the newly upgraded Wen Xiaoyan, which is equipped with an end-to-end phonetic language model and has become an emotional companion and all-round assistant.

First of all, Wen Xiaoyan has integrated information query including weather query, calendar query, unit conversion, stock price and so on.38 vertical assistantsIt can be seen that the efficiency of voice interaction is much higher than that of text interaction in these special scenarios.

Secondly, Wen Xiaoyan is rightProblems of timeliness and non-timelinessCan cope with encyclopedic inquiry, timely question and answer content of current political knowledge, and Wen Xiaoyan can search in real time, and follow the instructions accurately to reduce hallucinations; Non-timeliness issues such as common sense questions and answers are not a problem.

Finally, it is also the biggest difference between voice interaction and text. Wen Xiaoyan can be carried out with users.Emotional and natural communication, and can respond to feedback quickly, and realize realistic and anthropomorphic interaction effect.

Not much to say, let’s take a look at the actual effect of Wen Xiaoyan.

A major difficulty in speech recognition is dialect recognition. The pronunciation characteristics of dialects are rich and varied. The same dialect may have different pronunciations in different regions, and even the same word may have different pronunciations in different contexts. This makes it difficult for speech recognition system to accurately capture and analyze all pronunciation variants, which increases the difficulty of recognition. At present, Wen Xiaoyan has been able to deal with the characteristic dialects in Chongqing, Guangxi, Henan, Guangdong and Shandong. He can not only understand them, but also reply with the corresponding dialects.

Another major feature of voice communication is the need for multiple rounds of interaction. For example, in the following example, Wen Xiaoyan not only gives a method to distinguish budgerigars from different dimensions of adulthood and childhood, but also gives correct feedback in time when users interrupt to ask new questions.

Even if the reply contains many elements such as distinguishing the budgerigar’s sex and distinguishing from a certain fixed feature, Wen Xiaoyan gives a concise answer, and at the end reminds the user that he can record while observing.

In addition, human-computer interaction is often interrupted in the middle, such as the user gets the core information he wants or is not satisfied with the current output content, etc. When the user interrupts the voice playback to input, the voice recognition system may be mistaken because of environmental noise, unclear pronunciation of the user or confusion with the previous voice content.

In the face of children’s repeated interruptions, Wen Xiaoyan can accurately identify their needs of "changing a story", and when children say "Mom has already told it", they don’t automatically choose to change the story, but give an emotional reply in time to create an atmosphere of natural dialogue.

Wen Xiaoyan, an emotional interactive form, also extends its application to companion scenes in assistant scenes such as knowledge quiz. When the user mentioned "I’m in a bad mood", Wen Xiaoyan’s voice was worried and so on, and guided the user to tell the reason why he was in a bad mood and further enlighten him.

Second, brand-new cross-related attention, to create a very low training cost advantage

Different from the language model, the core difference of the phonetic language model is that it can produce emotion.

Jia Lei said that the text model only produces words, while the phonetic language model can have emotions. The key is two special links in the architectural diagram of the phonetic language model.TN Rhythm and Emotion Control of Personality and StyleThis is prepared for speech synthesis, which allows the big model to have the emotion of adapting the content while generating the answer, which is also the key innovation of Baidu’s end-to-end speech model.

Specifically, there are four key innovations.

First of all, this is released by Baidu.The industry’s first cross-modal speech language model based on Cross-Attention; The second is the model.Combining Encoder with speech recognition, the calculation of KV is reduced to 1/10.; The third isEncoder is combined with speech synthesis, and the output content can be controlled emotionally.; Finally,Efficient full query attention EALLQA, so that the KV cache is reduced to a few tenths.

On this basis, the model realizesIntegration of text recognition and text synthesis.On the basis of systematic end-to-end communication, these coupled technologies enable the model to realize a natural, realistic and emotional interactive experience on the basis of quick question and answer and quick understanding.

Jia Lei explained that the acoustic model is also a phonetic model, but usually large language models are connected by words. Therefore, in the process of integrating speech recognition and large language model, researchers share the process of Encoder and speech recognition in large language model to achieve the purpose of reducing the hard delay of speech interaction, and innovatively introduce cross-modal modeling, switching from Self-Attention to Cross-Attention, thus completing the integration of speech recognition and large language model.

Baidu proposes to use Cross-Attention to solve the cross-modal problem of speech and language. In this process, due to the speed limitation of the existing Attention technology in Cross-Attention speech language modeling, Baidu developed the EALLQA technology suitable for Cross-Attention, adopting implicit RNN two-level position coding, training MHA in 128 space, and inferring MQA in 512 space shared by all layers of the model, so as to achieveMake full use of limited training resources and reduce reasoning cost..

In the basic training of the model, Baidu uses self-distillation post-train the end-to-end speech language model of Cross-Attention based on the pre-training model of Self-Attention mature Wenxin language.

In fact,In speech model, the pressure of KV cache and KV calculation is much greater than that of text model.. Jia Lei explained that the essential difference between speech recognition and text model is that the first token at the beginning of a sentence determines the delay of speech recognition. For the large text model, users can wait for 2-3 seconds to give an answer after entering a paragraph of text, while in the large speech language model, users are less tolerant of reply delay, and they hope to hear the answer within 0.5-1 second.

On this basis, the end-to-end speech language model realizes low-cost training and low-cost high-speed reasoning. In addition, the speech language model also needs quick response and emotional response, which is the scene where another key technology is exerted.Streaming word-for-word LLM-driven multi-emotional speech synthesis. Jia Lei said that many rounds of continuous communication with emotion can make people want to continue communication.

Based on the streaming word-for-word method, its speech synthesis is to see one word jumping one word at a time, and the large model can help the speech synthesis to output the text normalized output, prosodic pause output and emotional output it needs, making the speech synthesis process flow like when people speak, and its adaptive emotional coverage can reach 17 kinds according to the text output.

In addition, there is a big pain point in speech recognition, which is that it can’t judge the starting point and ending point of the user’s speech, butLarge model blessing can make it analyze whether the user’s words have ended based on semantics.The semantics are incomplete and we need to wait.

Jia Lei further explained that,The activation of voice scenes requires extremely interactive cost, extremely fast delivery speed, intelligent and emotional humanized question and answer.. Baidu integrates speech recognition with large model to solve the problems of pre-fetching, hesitation, content understanding and quick question and answer, and integrates text synthesis with large model to output prosodic emotion needed in speech summation, thus solving the problems of context understanding and emotion control in synthesis. This makes the application potential of voice scene greatly improved.

Third, directly attack the difficulty of voice interaction, and enlarge Baidu’s end-to-end voice language model.

The continuous optimization of large model has significantly improved the robustness, naturalness and speaker similarity of speech, butThere are still many pain points in the previous technical path.This is why Baidu focuses on the end-to-end speech language model.

Compared with the communication between people, the response speed of large language model is slow, and users need to wait for a while to get a reply. In addition, voice communication is often accompanied by multiple rounds of dialogue and interaction, and it is extremely difficult for models to complete colloquial multiple rounds of interaction. Compared with texts, users use more scenes of voice interaction, and the surge of interaction will lead to the increase in the application cost of large-scale models, and the difficulty of large-scale application popularization will also increase.

On the traditional voice interaction route, it will be limited by the accurate response between contextual memory, noisy occasions, hesitant questions and interruptions.

Therefore, this has become a core contradiction in the field of voice interaction.The convenience of voice interaction determines its potential for large-scale application, and these pain points are hindering its popularization.Jia Lei believes that,The chemical reaction between speech and text is the key to find a breakthrough in a specific field in the future..

The appearance of phonetic language model is qualitative change, and its innovative synthesis technology makes the model not need to see the whole text of a sentence, but can synthesize a word by seeing the text of a word. On this basis, Baidu has excavated a unique application scenario. He gave an example. For example, when asking about the weather, users can quickly interrupt the next question when they get the temperature range of the weather. Its advantage is that the use cost of the model is greatly reduced, while the text model needs powerful hardware to achieve such efficient application, but the speech language model can achieve efficient concurrency with low-cost hardware.

At the same time, from the perspective of the whole voice interaction field, the accuracy of the speech recognition part of the large model has been greatly improved. Jia Lei believes that it is more a competition of speed, cost and answer accuracy.Cost reduction is the key to large-scale use of multimodal voice interaction..

Jia Lei said: "Cost reduction is the inevitable way of technological progress." The extremely low cost of Baidu’s speech language model means the possibility of large-scale industrialization. The landing application of AI is the core of the development of the model industry in 2025, and this model is the key to solving the speech problem.

Baidu’s accumulation in the field of speech recognition has a long history.

In 2018, the Deep Peak 2 model released by Baidu Voice broke through the traditional model that has been used for more than ten years, greatly improving the recognition accuracy in various scenarios. At the beginning of 2019, Baidu voice technology team announced the world’s first streaming multi-level truncated attention model SMLTA in the field of online voice, with a relative accuracy rate of 15%. In 2021, Baidu released the streaming truncated conformer modeling technology based on historical information abstraction-SMLTA 2, which solved the problems faced when the Transformer model was applied to online speech recognition tasks.

These technological innovations have been applied in many fields such as automobiles, consumer electronics and mobile phones. This time, in order to promote the large-scale application of speech language model, at present,Baidu has launched its online Xiaoyan and opened it for free, and plans to launch it on the open platform in April, and then it will be connected to call centers, speakers and other business lines..

Jia Lei mentioned: "Science may have national boundaries, but it has no corporate boundaries." Subsequently, Baidu opened the end-to-end speech language model, just to promote the application of the big language model in the speech field, which is conducive to the development of the whole industry and ecology.

Conclusion: Baidu’s phonetic language model is unsheathed, opening a new chapter with low cost and high efficiency.

Convenient, efficient, natural and friendly voice interaction, which is widely used in multiple scenes, is very important in the digital age. However, from the current actual effect, voice interaction has greatly improved in recognition accuracy, and now the competition focus of the industry is on its recognition speed, cost and accuracy of answers.

In this context, the release of Baidu’s end-to-end speech language model further broke through the cost, and proposed a new technical path, which pushed the industry competition of voice interaction to a new height. At the same time, Baidu will launch it on an open platform, which will accelerate the application and popularization of large models in voice interaction scenarios.

　　BEIJING, Beijing, May 10 (Xinhua)-"44% of fraudulent short messages come from paragraph 170/171". In the past month, the serious problem of telecom fraud in paragraph 170/171 in virtual operator has aroused great public and official concern. “‘ 170/171’ Do you dare to answer the phone at the beginning? " Worried, the Ministry of Industry and Information Technology, after interviewing relevant enterprises, recently issued a document asking virtual operator to complete the work of re-registration of user identity information within one month. According to a survey conducted by a reporter from Zhongxin. com (WeChat WeChat official account: cns2012), at present, real-name registration system, virtual operator is facing a series of technical screening problems, and the phenomenon of non-real name sales still exists, but the real-name registration system degree of "170/171" has been improved.

　　No.170/171 Duan Cheng Telecom Fraud New Label Users are afraid to answer when they see "170/171"

　　Tencent’s mobile phone housekeeper security experts found through big data monitoring that "nearly 44% of fraudulent SMS messages came from virtual operator’s 170/171 paragraph." Behind this data is that paragraph 170/171 has become a new label of telecom fraud. In the daily monitoring, the Anti-telecom Fraud Center of Guangzhou Public Security Bureau found that the numbers starting with 170 and 171 have become high-risk telephones for telecom fraud.

　　In the past month, with the risk warnings issued by the Ministry of Industry and Information Technology and the public security departments, as well as the media and WeChat friends circle’s reports and publicity on the fraudulent practices of paragraph 170/171, the public’s vigilance against paragraph 170/171 fraud has been greatly improved.

　　Mr. Chen, who is engaged in the publishing industry in Beijing, told the reporter of Zhongxin.com, "There are two friends around me who have been cheated by the fraudulent phone call in paragraph 170. Now I dare not answer the phone call in paragraph 170, let alone click to view the text message sent by paragraph 170."

　　"Excuse me, do you need a loan? Do you need to invest in financial management? … …” Through investigation, the reporter found that paragraph 170/171 was used by fraudsters and also by some enterprises for marketing, which became annoying phone calls and text messages for the public. Several Beijing taxi drivers and supermarket shopping aunts interviewed randomly by reporters all said, "Recently, the number of calls received from various investment and wealth management or some commodity promotion in paragraph 170 has increased significantly, and I don’t know how to know my number."

　　Fraud takes advantage of the victim’s greed and cheapness to pretend to be a special subsidy issued by the government.

　　According to the fraud alert data released by the Anti-telecom Fraud Center of Guangzhou Public Security Bureau last week, there were 93 telecom fraud alerts involving paragraph 170 last week, among which the most prominent fraud methods were pretending to be government departments to issue special subsidies and pretending to be teachers to say that children needed surgery fees for injuries. There were 87 cases of telecom fraud involving paragraph 171, of which 67% were fraudulent by sending short messages with Trojan links.

　　According to Zhongan Online, at 19 o’clock on April 25, Mr. Feng suddenly received a text message: "××××, your child’s performance at school has dropped. Please click the link below to check the score." Mr. Feng looked at the number that sent the short message. It was a number that started with 171 and belonged to Hefei. Mr. Feng casually clicked on the link attached to the short message, but I didn’t expect it to open. Not long after, the bank card bound by the mobile phone sent a reminder message saying that 2000 yuan in the card was taken away.

　　According to reports, the telecom fraud in paragraph 170/171, pretending to be a government department to issue special subsidies, pretending to be a teacher to say that children need surgery fees, credit card withdrawals, handling large credit cards, and neonatal subsidies, all take advantage of the victim’s greed for benefits.

　　The police reminded the general public that when receiving a phone call or text message at the beginning of such a number segment, they should verify the real identity of the other party, don’t click on the link in the text message easily, don’t transfer money or make online transactions casually according to the requirements of the other party, and install mobile phone security protection software as soon as possible to prevent being cheated.

　　When the official rectification is in progress, the real-name registration of section 170/171 should be completed before the end of May.

　　In fact, on April 7th, the Network Security Administration of the Ministry of Industry and Information Technology, the Information and Communication Development Department, and the Information and Communication Administration, together with the Beijing Communications Administration, urgently interviewed three mobile communication resale enterprises (virtual operator) that were not implemented in real-name registration system. The three companies are required to immediately conduct a comprehensive investigation on the development of the company’s real-name registration system and seriously carry out rectification.

　　On April 27th, the Ministry of Industry and Information Technology further issued the Notice on Strengthening Standardized Management to Promote the Healthy Development of Mobile Communication Resale Business, further requiring resale enterprises, that is, virtual operator, to immediately carry out self-examination and self-correction, and complete the work of re-registration of user identity information for telephone numbers that were not registered in real name or falsely registered in the previous period within one month.

　　At the same time, the notice made it clear that resale enterprises should strengthen the supervision and management of entity and network agency channels, strictly enforce the access conditions of agency channels, distribute uniform signs for agency channels, and prohibit the signing agency channels from developing subordinate agents without authorization. If a resale enterprise goes through the network access formalities through physical marketing channels, it must verify and automatically enter the user identity information through the second-generation ID card identification equipment, and it is forbidden to manually enter the ID card information. If you go through the network access procedures, you should verify the user’s identity in strict accordance with the relevant requirements. Resale enterprises should establish a regular self-inspection system for real-name registration and increase the inspection of their own marketing and principal-agent channels.

　　Virtual operator: Rectifying and benefiting the development of the industry faces technical screening problems.

　　According to the public information of the domestic virtual operator platform, since April, most virtual operator have started self-examination and self-correction, and the real-name registration system situation of the virtual number segment has been improved to some extent. A president of virtual operator, who asked not to be named, told the reporter of Zhongxin.com on the 9th that the work in real-name registration system is related to the immediate interests of users and the healthy development of communication enterprises and industries. The company has raised the standards of rectification and further strengthened the implementation of the work in real-name registration system.

　　However, the process of strengthening real-name registration system in virtual operator is facing the difficult problem of technical means screening. According to an employee of a marketing department in virtual operator, the second-generation ID card reader distributed by the company to offline channel vendors has been cracked, and it is currently working with relevant enterprises and departments to make up for the leak.

　　When the reporter visited a newsstand selling virtual number segments in Beijing, the newsstand sales staff said that they could help to handle a virtual mobile phone number of segment 170 without issuing an ID card, and they could use other people’s ID card information to help. A person in virtual operator said that some vendors used other people’s ID card information in violation of regulations, taking advantage of the "One Certificate No.5" loophole. Virtual operator needs to strengthen the inspection of channels together, screen through technical means, identify users by telephone call back for verification, and clean up fraudulent numbers in time.

　　Real-name authentication from online sales channels is also difficult. Although some virtual operator online real-name registration systems are enabled, it is required to fill in the user’s real name and ID number, and upload the front, back and half-length photos of my second-generation ID card. However, according to the IT Times, some virtual operator registered with fake real names can still enter the order submission and payment process smoothly. In the whole process of online real-name authentication, we mainly rely on manual audit and express copy of ID card, and there will be more uncontrollable situations. Zou Xueyong, secretary-general of China virtual operator Industry Alliance, thinks that technical means of authentication, such as video face recognition, should be strengthened.

　　American cars cost oil, and this old-fashioned impression will be completely changed with the launch of Lincoln electric hybrid models. On March 6, Lincoln launched the Lincoln electric hybrid family composed of the new electric hybrid Lincoln Z, the new electric hybrid Lincoln Adventurer and the electric hybrid Lincoln Navigator in Shanghai, redefining the luxury consistency standard with the three core product values of inductive acceleration, fuel consumption halving and worry-free battery life, and achieving the "first choice of luxury hybrid". It also announced that the new electric hybrid Lincoln Z and the new electric hybrid Lincoln Adventurer were officially launched. The two major products of Lincoln Z and Lincoln Adventurer were also updated to 2024 models across the board.

　　Among them, 2024 Lincoln Z launched 2 models, priced 235,800 – 254,800 yuan; 2024 Lincoln Adventurer launched 4 models, priced 235,800 – 345,800 yuan. Electric hybrid, Lincoln Z electric hybrid version provides 3 models, priced 274,800 – 365,800 yuan; Adventurer electric hybrid version launched 2 models, priced 253,800 – 273,800 yuan.

　　The new electric hybrid Lincoln Z fuses a unique American DNA and "Z aesthetic" design. In addition to the car colors used in the models on sale, the new electric light silver exclusive car colors, and electric hybrid exclusive 18 and 19-inch wheels are available. With the blue Lincoln star logo, blue model nameplate and other exclusive design details, two new theme interiors, "Quiet Blue" and "Autumn Chestnut Brown", have been added to the cabin, bringing a new pleasant sensory experience with both trend and luxury, showing the unique identity of the new electric hybrid Lincoln Z.

　　In terms of power, the new electric hybrid Lincoln Z comes standard with a 2.0T engine, which can output a maximum net power of 214kW and a maximum torque of 410N · m. On this basis, an electric drive system is added to drive the motor to a maximum power of 140kW and a maximum torque of 320N · m, achieving a 6-second 100-kilometer acceleration.

　　The new electric hybrid Lincoln Adventurer adopts the new Quiet Flight 2.0 design concept, adds the exclusive car color of electric light silver, and uses the same color surrounding parts as the fuel version of the high-end model. Unique bright gray decorative parts and exclusive design details such as the Lincoln star logo on the blue background, the side nameplate of the blue model and the new theme interior of "Quiet Blue" show a sense of respect.

　　In terms of power, the new electric hybrid Lincoln Adventurer is equipped with a 1.5T engine with a maximum torque of 226N · m, and the comprehensive power of the system reaches 153kW. With the E-CVT electronically controlled stepless gearbox, it can calculate the optimal efficiency of the engine and motor in real time, automatically match the power output mode, achieve more powerful and efficient power output, and easily deal with various working conditions on the road.

　　In terms of performance, both the new electric hybrid Lincoln Z and the new electric hybrid Lincoln Adventurer use Power Split power shunt hybrid technology. At the same time, the dual motors of the Lincoln electric hybrid family work together to achieve up to 100% braking energy recovery. With eight-layer flat-wire motors, the system efficiency can reach up to 96.5%. The efficient hybrid mode keeps the engine in the best working condition at all times. With the support of the advanced electric hybrid system, the fuel consumption of the new electric hybrid Lincoln Z is as low as 5.99L under the comprehensive working condition of 100 kilometers. The fuel consumption of the new electric hybrid Lincoln Adventurer is as low as 5.35L under the comprehensive working condition of 100 kilometers, and the fuel consumption in urban working conditions is as low as 4.71L/100km. Compared with the fuel version, the fuel consumption is reduced by 52%.

　　It is worth mentioning that the electric hybrid system of the two new cars is equipped with a new generation of ternary lithium battery cells with a safe and reliable integrated water cooling design, which does not need to be charged through an external charging pile. It is not afraid of cold and heat, does not lose power at low temperatures, and does not catch fire at high temperatures. While cooperating with the electric drive system to work efficiently, it effectively solves the anxiety of pure electric mileage. The new electric hybrid Lincoln Z has a battery life of up to 885 kilometers, and the longest range of the new electric hybrid Lincoln Adventurer can reach 1252 kilometers. No matter whether it is a daily car or a long-distance self-driving trip, there is no need for frequent charging and replenishment. It is far away from the performance troubles caused by insufficient power, bringing a "zero anxiety" all-scene consistent luxury driving experience.

　　In terms of intelligence system, the new electric hybrid Lincoln Z and the new electric hybrid Lincoln Adventurer are equipped with 12.3-inch full LCD meters and a 27-inch through-type high definition panoramic screen with a resolution of 4K level, the car system adopts Qualcomm Snapdragon 8155 chip, the new electric hybrid Lincoln Z is equipped with a new generation of Lincoln Co-Pilot 360 ? 2.1 intelligent driving assistance system, with Lincoln BlueCruise active driving auxiliary features, and also integrates adaptive cruise, lane centering, intelligent speed limit and other functions, which can realize L2 + level intelligent assisted driving.

　　At the same time as the launch of the new electric hybrid Lincoln Z and the new electric hybrid Lincoln Adventurer, the two major products of Lincoln Z and Lincoln Adventurer have also been refreshed to 2024 models across the board. Among them, the "Smart Leader New Generation Car" Lincoln Z car series continues to "up the ante" in terms of intelligence, luxury and safety, bringing a dynamic smart travel "Z experience" to meet Generation Z’s pursuit of independence and individuality. The 2024 Lincoln Adventurer Corsair leads the pure American luxury experience by leveraging its unique four core product advantages of new American luxury ceremony, strong power of leap, leap luxury comfort, and leap safety.

　　(Xinmin published this article for the purpose of transmitting automobile information, which does not represent the views of this website)

　　[October 27, 2023, Beijing] Today, Extreme Krypton Intelligent Technology officially released the pure electric hunting super run — — Krypton 001 FR, the official retail price starts at 76.9 yuan. Extreme Krypton 001 FR combines the performance of tens of millions of super-running, the extreme safety of "highway", the luxury and comfort of executive cars, the ability of off-road SUVs to get rid of difficulties, as well as the top intelligent cockpit and intelligent driving experience, the driving ability of F1 world champions, and the car satellite phone that never loses contact, which subverts the limits of traditional super-running and brings users real super-running in the era of intelligent electric power.

　　Pure electric hunting super running pole Krypton 001 FR official retail price

　　Users who decide before December 31st can not only enjoy the six-year or 150,000-kilometer vehicle warranty, three-power lifetime warranty, free data flow, free road rescue during the warranty period, and free 7kW smart home pile-filling package (50 meters), but also get a "Raikkonen model" worth 60,000 yuan, with the same adjustment and driving experience as the "F1 champion".

　　Extreme performance continues to evolve, with an acceleration of 2.02 seconds and a maximum horsepower of 1300 horses.

　　Since its debut, Krypton 001 FR has been launched with five core technologies in global mass production: the four-motor distributed electric drive launched in global mass production, the world’s first ZVC four-wheel torque vector control, the world’s first "dragonfly structure" middle-section integrated die casting, the world’s first 8295 intelligent cockpit computing platform and the world’s first mass production satellite communication technology, making it the strongest pure electricity on the surface in the era of intelligent electric power.

　　In just over a month, Krypton 001 FR once again subverted the limit and continued to evolve. Extreme krypton’s pursuit of the ultimate makes its acceleration from 2.07 seconds to 2.02 seconds, and it only takes 6.29 seconds to accelerate from 0 to 200km/h, which is more than half a second ahead of the tens of millions of super-runs. Under the continuous adjustment of the engineering team, the 001 FR has added a special "ejection mode", which makes its starting time 0.02 seconds faster than the normal acceleration. With it, the 001 FR straight-line racing can always be one step ahead of others.

　　The 0-200km/h acceleration of Krypton 001 FR takes only 6.29 seconds.

　　The intelligent drive evolution of Krypton 001 FR is not only an instantaneous explosion, but also a full-range high-energy power output. The "overclocking mode" developed by Krypton for 001 FR turns the "nitrogen acceleration" function in racing games into reality, making the maximum horsepower burst instantly. The power of overclocking mode is increased by 10% compared with that of standard mode at high speed, and it can be started at any time of acceleration, so that the extreme krypton 001 FR can maintain a strong explosive force during the whole acceleration, whether it is cornering acceleration or overtaking straight. After the overclocking mode was started in "Raikkonen mode", the maximum horsepower of Krypton 001 FR was also increased from 1,265 at the time of debut to an astonishing 1,300.

　　Relying on Raikkonen mode and overclocking mode, Krypton 001 FR racing shows full-range high energy.

　　Behind the extreme performance of Krypton 001 FR is the core technology of Krypton high-efficiency hard core. Extreme Krypton 001 FR is equipped with the world’s first mass-produced 800V high-voltage system of silicon carbide, and each motor is equipped with a silicon carbide electronic control module separately, which is stable, efficient and accurate in control. The combined output power of the four motors can reach up to 956kW, matching with the leading performance of 100kWh Kirin battery, and the maximum charging rate can reach 4C. It only takes 15 minutes to charge the battery from 10% to 80%. The discharge power of krypton 001 FR battery reaches megawatt level. At the same time, the "straight waterfall oil cooling technology" is used to dissipate heat for the motor, and the PTM 2.0 global thermal management system is used to dissipate heat for the battery. The powerful thermal management system allows the peak power of krypton 001 FR to be continuously output for 12 times without attenuation. After 8 consecutive times of uninterrupted acceleration from 0 to 100km/h, and then braking to 0, the internal temperature of the battery only increased by 3.6 degrees during the vehicle test.

　　It is not only the temperature, but also the strength of the Krypton 001 FR intelligent electric drive that keeps stable. Krypton uses a 249-meter-long high-strength carbon fiber material to cover the motor rotor, which is covered with 15 layers. The sleeve is only 1.1 mm thick and weighs only 88 grams. It can withstand more than 16 tons of tension at the highest speed. This is also the reason why the maximum speed of the rear double motor of Krypton 001 FR exceeds 20,000 rpm, and its high-speed performance can still explode to the maximum speed.

　　Extreme safety is the strongest confidence on the surface, delivering 160 thousand, the only zero spontaneous combustion in the industry

　　Delivered more than 160,000 units, the only zero spontaneous combustion record in the industry, and the safety limit of electric vehicles has been constantly refreshed. Safety is the core gene of krypton, and krypton safety is the foundation of extreme performance of krypton 001 FR. Extreme Krypton 001 FR has been specially upgraded on the basis of the technology of "Road Tankryon". In addition to the 720-degree omni-directional safety armor, the middle part of the body of Extreme Krypton 001 FR adopts the world’s first "dragonfly structure" middle-section integrated die casting, and the entire aluminum rear body is also manufactured by the world’s original "four-section" rear-end integrated die casting process to minimize the impact on the battery pack during collision. Inside the battery, Krypton 001 FR adopts the world’s first double-large-face liquid cooling technology, and the heat exchange area is four times larger than that of the traditional liquid-cooled battery pack at the bottom, which realizes the integration of three functions: cooling, heat insulation and structural strengthening, and adds code layer by layer to effectively eliminate fire hazards.

　　Extreme krypton safety comprehensive advanced

　　"Not afraid of collision and not catching fire" is only the basis of extreme krypton safety, and excellent control is the core of extreme krypton 001 FR safety. Relying on ZVC four-wheel torque vector control, Krypton 001 FR can not only realize the tank turning without picking the road, but also realize the high-speed turning of the track. In the track test, the speed of Bugatti crossing S-bend on dry ground is 80km/h; Sprinkling water in the same bend, extremely Krypton 001 FR can cross the S-bend at a high speed of 80km/h, with stable body and precise control. Based on this, Krypton has also brought the much-anticipated "drift mode", so that drivers can enjoy the exciting high-speed drift and cornering experience.

　　Excellent performance under extreme dynamics benefits from the perfect cooperation of top software and hardware. Krypton 001 FR has the world’s largest carbon fiber roof, which is 1.46m long and 1.16m wide, and weighs only 6.5kg, which is 65% lighter than the traditional glass roof and 67% stronger. In addition, the front lip, side skirt, tail and diffuser of Krypton 001 FR are all made of carbon fiber aerodynamic kits, which can reduce the overall weight by more than 17kg and increase the downforce by 178kg when the vehicle is driving at high speed. With the strong grip brought by high-performance tires with narrow front and wide back at the track level, you can always maintain the best "flying on the ground" posture no matter how fast you are. KW, the world’s top shock absorber brand, customized the shock absorber for the original factory of Krypton 001 FR, so that the vehicle can perfectly match different track characteristics and different driving styles, bringing a more pure dynamic texture. The original factory comes standard with AP Racing 10-piston forged aluminum brake calipers and racing car’s exclusive oversize Brembo carbon ceramic perforated brake disc, so that polar krypton 001 FR rushes into the two-second club, while the braking distance of 100-0km/h is only 33.4 meters, even if it is driven fiercely for a long time, it can maintain the best braking state.

　　Excellent control is the core of kryptonian security.

　　The global evolution Raikkonen model allows you to have the same driving experience as the F1 champion.

　　The evolution of krypton is always inseparable from the co-creation of users, and the evolution of krypton 001 FR ushered in the help of F1 world champion kimi raikkonen. As the first European car owner and the chief performance consultant of Krypton 001 FR, kimi raikkonen participated in the development and adjustment process of Krypton 001 FR, and constantly tapped the great potential of this performance beast. At the press conference, An Conghui, CEO of Extreme Smart Technology, and kimi raikkonen jointly started, and the world’s first world champion who used intelligent algorithms to write in the car — — Raikkonen model copies the experience and ability of F1 champion to all owners of Krypton 001 FR.

　　The experience and technology of F1 world champion Raikkonen will be perfectly integrated into krypton intelligent drive.

　　The rich track experience and skillful skills of F1 world champion are perfectly integrated into the intelligent drive system. Raikkonen mode not only brings the maximum horsepower increase of 1,300 horses to Krypton 001 FR, but also makes exclusive customized adjustments to nine parameters of Krypton 001 FR, giving users a real chance to have the same driving experience as F1 champion. The new model will be officially launched by OTA in the first quarter of next year after the winter test. In the future, the evolution of the extremely krypton intelligent drive system will also empower all extremely krypton 001 models to achieve advanced performance.

　　The 8295 intelligent cockpit computing platform, which was first launched and delivered in the global mass production of extreme intelligence.

　　As a hard-core technology-driven company, Krypton has the ultimate pursuit for every product. Extreme Krypton 001 FR not only has the ultimate safety and performance, but also completely subverts the industry’s cognition of super-running and high-performance vehicles. Extreme Krypton 001 FR is the first in global mass production and the first to deliver the 8295 intelligent cockpit computing platform, which is 2.2 times more powerful than the 8155 computing platform, and the graphics and video processing capacity is 2.7 times higher. Based on the 8295 computing platform, Krypton has created an exclusive intelligent space for pure electric supercar for 001 FR.

　　Krypton created the world’s first exclusive intelligent space for pure electric supercar for 001 FR.

　　The intelligent follow-up seat solves the problem that the traditional bucket racing seat sacrifices comfort for performance. Daily driving can provide 10 points and 6 massage functions, and the comfort is full. When driving fiercely, the seat will respond dynamically within 0.1 second according to the vehicle dynamics, increasing the lateral support of the waist and legs, so that the driver can realize the integration of people and vehicles in the most comfortable sitting position. In addition, Extreme Krypton 001 FR will also be equipped with a brand-new 2.5K flexible OLED central control panel, 35.5-inch super-large imaging HUD, competitive style instruments and atmosphere lights as standard; There is also the top Yamaha stereo, with a rated power of 2400 watts and 28 speakers in the whole car, so that each seat is 7.1.4-channel surround sound. Top-level full-featured attributes also satisfy the passion of track driving and the comfort of street commuting.

　　In addition to hardware, there is also a comprehensive upgrade of software. The full digital capability of the krypton intelligent drive system opens all the performance parameters of the krypton 001 FR. Extreme Krypton 001 FR matches the first built-in professional track APP in the history, which can make all four categories of performance parameters open, each item can be adjusted by more than 10 gears, and there are more than 13,000 custom driving styles. In addition, this track APP also has a perfect video recording function, which can perfectly restore the driver’s driving performance on the track and easily share it with other riders. It is a real track dating artifact.

　　In terms of intelligent driving that users pay attention to, Extreme Krypton 001 FR is also an intelligent pure electric supercar that can be equipped with laser radar, and has intelligent driving ability covering the whole scene of cities and high speed. Lu Yu extreme weather, strong perception can still better identify road conditions, both safe and worry-free. Even if we drive to no man’s land, the vehicle-mounted satellite communication technology, which was launched in the global mass production of Krypton 001 FR, can escort users and make safety never lose contact.

　　Krypton 001 FR is an intelligent pure electric supercar that can be equipped with laser radar.

　　Embracing the era of users, polar krypton provides sufficient choice space for car owners. All car owners of polar krypton 001 FR will have two kinds of car roofs for free: if you want to be extremely pure, you can choose the world’s largest carbon fiber car roof; If you want to be intelligent and advanced, you can choose lidar+intelligent light roof. No matter which roof you choose, the ZAD fully intelligent driver assistance system will be standard on the Krypton 001 FR, which will bring users an unprecedented leading intelligent experience in super-running.

　　Let the face value of hunting clothes, the performance of super-running, the comfort of cars, the space and ability of getting rid of difficulties of SUVs, and the top-level intelligent experience be met at the same time in one car. Get on the track, get off the street, and even pursue poetry and distance. Extreme Krypton 001 FR will be an all-round pure electric hunting supercar covering all road conditions and scenes.

　　Full road condition, full scene, all-round pure electric hunting super run

　　Eight hard-core technologies were launched in the global mass production, and six performance parameters ranked first in the global mass production, with over 100 luxury configurations as standard. Extremely, 001 FR once again subverted the tradition and opened the era of intelligent electric supercar. Whether it’s hunting coupe or hunting supercar, Krypton has always evolved in the spirit of pursuing the ultimate, setting a real value benchmark for users.

　　-End-