Music Radar: the technology into business[i world network operators Note With the development of technology, consumers are Nike Hyperdunk 2014 increasingly diverse needs can be met. For example, the sound mentioned herein search is such a music search to meet consumer demand for products that not only provide matching soundtrack by copying the search, Nike SALOMON Air Jordan 6 Women but also support from the human voice singing search. Text / world network operators when you driving down the road when suddenly some familiar melody came from the radio, you can not help but want to get it out, Nike Air Jordan 11 Women but can not remember its name. This familiar scene in Lei's business idea, he was writing for a solution is this: open the phone, called a musical radar applications, facing it hum a familiar melody, the system will be based on melody Go to the database search, and complete music presented on the screen. This kind of melodic music search function is not mysterious, more domestic and foreign Internet Nike Air Max 2012 Men giants have the layout. And recently, micro-letter shake quietly on the line, Nike Air Jordan 5 Women putting the soundtrack music search features to make music search imagination further amplification. Music radar unique place in that it not only provides matching soundtrack by copying searches also support from the human voice singing search. Music Radar founder Wang Lei told the 'world network operators \u0026 middot; managers' correspondent, compared with soundtrack search, humming search in the technical difficulty is much higher. Only the former can be found in the database, and some fragments of the same music, most need to consider noise or rhythm of the pace, but the latter is in addition to noise interference suppression, but also need to be treated for the user out of tune, pitch allowed. \u0026 Ldquo; the former is precise search, which is the fuzzy search. \u0026 Rdquo; he said. Bidirectional layout In 2006, Chinese Academy of Sciences Institute of Automation Du Bo Wang Lei came into contact with the voice recognition in this field, in the laboratory environment to match the human voice and the database. That time, not yet have mobile Internet environment, and in the music search, identification and interaction in the field has not yet appeared complete business model. Faced with a mature market yet, Wang Lei chose dormant. Let Wang Lei opportunity to generate entrepreneurial impulse from abroad of a company called Shazam. 2011, make music search technology Shazam not only go through the line, but also in one fell swoop through the business model. In simple terms, Shazam is similar to a music distribution platform to guide users to download music through search, and then pay for. Wang Lei found the year 2011, Apple's music store seven or eight billion in sales, 10% is contributed by Shazam. This makes him extreme emotion: Good Times music search market has come. However, compared to the foreign environment, whether in Western Europe or North America, the proportion of users of credit cards are quite high, while domestic product to pay for virtual environment is not mature, simple copy Shazam model, feasibility apparently not strong. It is based on this consideration, Wang Lei radar this product only for the music chosen way of walking on two legs, two-way layout 2B and 2C fields. Profit model, music radars on the role of B-line, mobile music base open access platform at the C-terminus, if a user searches on the music radar to good songs, will be able to directly pay to download and set as ring tones on the phone, and Music Radar can derive 25% to 30% of the SALOMON share. Wang Lei told the 'world Nike Air Jordan 5 network operators \u0026 middot; managers' reporter, currently paid CRBT subscribers mainly in the following three cities. In addition, the music radar also by embedding other music clients become technology providers, cool dog music, cool Cheap Nike Free SALOMON Outlet music and in my movement Migu music technology Nike Air Max on music search and music are integrated radar products. Technical principle from the principle point of view, music search belongs to a sub-speech recognition, based on \u0026 ldquo; deep learning \u0026 rdquo; theory \u0026 mdash; \u0026 mdash; this technology through the simulation of the human brain's behavior in a certain accumulation after, according to the algorithm identified corresponding content. Depth study of the earliest applications are in Google. Google has developed a deep learning machine without manual annotation, image input directly to the machine, a few months later, the computer will be able to find a similar or identical items from the input of the picture. Wang Lei told the 'world network operators \u0026 middot; managers' reporter, and most people understand different music search and language-independent, its main judgments based on melody rather than the lyrics. User on the client record and upload music radar of the server, back-office systems to which the melody first feature extracted to generate a time series Nike KD VI of a note, then, and database for quick match, find identical or similar to the sequence of time document, the final output to the client. Including technical standards based mainly on the accuracy and speed of response. Looking at the domestic and international market, currently singing voice search technology into commercial real still very few cases, in addition to music radar, only another foreign application SoundHound. By phone preloaded music radar has covered China 400 million users, and from the data, users from geographical distribution to age, identity and so are more dispersed. Specifically, the following three cities users often bring high-value user CRBT revenues, north of Guangzhou-Shenzhen and other cities users, few people will pay for the ring tones. Many of them use search to find a more familiar songs, or even just for fun, I want to know whether their singing out of tune and keep singing against App. Data accumulated from the current multiplication usage scenario, the user while driving, watching television is the most intensive use of the music scene radar App. In many places, the user is already using uncommon disco music radar, and this gives Wang Lei was quite surprised, because this scene was not in his vision being. Beyond the expected results contemplated, let Wang Lei cherish the accumulated data. \u0026 Ldquo; we will save the original recording, which sampled 1,000 daily manual monitoring, recording and Nike Air Max 2012 found a lot of disco music background. \u0026 Rdquo; Wang Lei to 'world network operators \u0026 middot; manager' told reporters last December 31, Hunan Satellite TV's New Year's Eve concert, also once let the music radar traffic soared. Data accumulation, better, Lei began to sketch the future: Music Radar will use the data to provide services for B-side. For example: Consumers a region like what kind of music, for what kind of advertising, these data bases to move to this music, it has immeasurable significance. And the music radar data involved in such a service, its first-mover advantage. Wang Lei cited an example: A user often listen to songs on the Caribbean Sea, the music radar can push its Caribbean tourism advertising; and listen to users Beatles, basically middle-aged people Reebok CL Chi Kaze with higher incomes, this Air Jordan Outlet category may is facing wards, transfer, music radar can push the car to its advertising. At the same time, with the data speak, in many cases it also brings convenience to teamwork. Prior to this, a singer Yangjiang City, Guangdong commissioned music for the songs do personal recommendation radar, radar music in the background statistics technical team found that users of the singer Yangjiang local search ranking very high. This collaboration then reached soon. From on-line in early 2013. So far, the music radar has been close to breakeven, and has established a database of 20 million songs. For Wang Lei, in addition to serving as good, but he also hoped that in the coming time, the music shopping guide steps towards commercialization faster. (This article first appeared in the i world network operators, all reproduced, please indicate the author and source)