CN104464754A - Sound brand search method - Google Patents
Sound brand search method Download PDFInfo
- Publication number
- CN104464754A CN104464754A CN201410752310.9A CN201410752310A CN104464754A CN 104464754 A CN104464754 A CN 104464754A CN 201410752310 A CN201410752310 A CN 201410752310A CN 104464754 A CN104464754 A CN 104464754A
- Authority
- CN
- China
- Prior art keywords
- trade mark
- audio frame
- comparison
- measured
- audio
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Auxiliary Devices For Music (AREA)
Abstract
The invention provides a sound brand search method. The sound brand search method includes the steps that sample brands are divided into two or more groups according to the time length; the group of the brands is determined according to the time lengths of to-be-measured brands to conduct similarity searching in the group; the to-be-measured brands and the sample brands are segmented into a plurality of voice frequency sequences equal in length; the total voice frequency frame numbers P in two voice frequency sequences, needing to be compared, of the to-be-measured brands and the sample brands are determined; the energy ratio J of each corresponding voice frequency frame in the two voice frequency sequences needing to be compared is calculated to judge either two corresponding voice frequency frames are normal voice frequency frames or abnormal voice frequency frames, and the average energy ratio K of the normal voice frequency frames in the two voice frequency sequences needing to be compared is calculated; the voice frequency frame number Q of the energy ratios in a section of the average energy ratio K is determined; the number ratio of the voice frequency frame number Q to the total voice frequency frame numbers P is calculated for each voice frequency sequence, and the similarity of the two voice frequency sequences is judged according to a specific value S of the number ratios of the two voice frequency sequences needing to be compared to judge the similarity between the to-be-measured brands and the sample brands.
Description
Technical field
The present invention relates to speech recognition retrieval technique, particularly, relate to a kind of sound trade-mark searching method.
Background technology
Trade mark is the mark of company, product or service, combines together, play very important effect, be an important attribute of company and products thereof, have uniqueness in industry and commerce society with the commercial quality of enterprise, service quality, operation control.For making trade mark obtain legal protection, must to trademark office's official registration.Along with the quickening of China's expanding economy and globalization process, trade mark quantity cumulative year after year.Prevent repeated registration or similar brand registration from being the key problem of trade-mark administration.
The execution from 1 day May in 2014 of new Trademark Law, regulation sound can be used as trademark application registration.Sound trade mark is the one of non-traditional trade mark, can be the same as the key element of trade mark with other, possess and by the product of an enterprise or service with the product of other enterprises or serve the basic function distinguished and come, namely can must have notable feature, be convenient to consumer's identification.Whether sound trade mark is similar, mainly be whether be enough to obscure judge to the relation between trade mark and commodity by the sense of hearing by consumer, to the two kinds of commodity employing similar mark, whether consumer can produce misidentification to the source of commodity when general attention power.
Therefore, in order to protect the legitimate rights and interests of registration sound trade mark, need to retrieve sound trade mark to be registered, compare with registered sound trade mark, determine that the two is not identical or be not similar, just have registration and qualification.
Summary of the invention
The object of this invention is to provide and a kind ofly can accurately identify audio frequency similarity thus ensure the accuracy of sound trade mark retrieval and the high sound trade-mark searching method of recall precision.
According to technical scheme of the present invention, provide a kind of sound trade-mark searching method, the method comprises the following steps: the sample trade mark in trade mark storehouse is divided into plural group according to duration; Obtain trade mark to be measured, and determine the group belonging to described trade mark to be measured according to the duration of trade mark to be measured; The similarity retrieval of trade mark to be measured is carried out: multiple tonic trains trade mark to be measured and sample trade mark being divided into equal length in determined group; Determine the audio frame sum P needed in trade mark to be measured and sample trade mark in the two-way audio sequence of comparison; Calculate the energy value needing each audio frame in the two-way audio sequence of comparison, with the energy ratio J of the audio frame of each correspondence in the two-way audio sequence obtaining needing comparison, judge to need two audio frames corresponding in the two-way audio sequence of comparison to be normal audio frame or anomalous audio frame according to the size of described energy ratio J, and calculate the average energy ratio K of the normal audio frame needed in the two-way audio sequence of comparison; Set the interval of described average energy ratio K, and determine to need the audio frame number Q of energy ratio within the interval of described average energy ratio K in the two-way audio sequence of comparison; For tonic train described in each road, calculate the quantity ratio of audio frame number Q and audio frame sum P, and judge the similarity of the two-way audio sequence needing comparison according to the ratio S needed between the quantity ratio of the two-way audio sequence of comparison; And the similarity of trade mark to be measured and sample trade mark is judged according to the similarity between the trade mark to be measured tonic train corresponding with sample trade mark.
According to technical scheme of the present invention, first the sample trade mark in trade mark storehouse is divided into groups, plural group is divided into according to the duration of trade mark, when carrying out similarity retrieval to trade mark to be measured, determine group belonging to it according to the duration of trade mark to be measured and retrieve in determined group, and without the need to retrieving in whole trade mark storehouse, shorten retrieval time, substantially increase trade mark recall precision.For similarity retrieval step, technical scheme of the present invention is by being divided into minimum comparison unit tonic train by the melody of sample trade mark and trade mark to be measured, and to form tonic train audio frame one by one energy comparison value (i.e. amplitude) whether similarly judge between the tonic train that trade mark to be measured is corresponding with sample trade mark, and then judge the similarity of trade mark to be measured and sample trade mark, and technical scheme of the present invention eliminated anomalous audio frame before audio frame comparison, only keep the comparison of melody, avoid the interference of other sound in trade mark, can accurately identify audio frequency similarity, ensure that the accuracy that sound trade mark is retrieved.
Other features and advantages of the present invention are described in detail in embodiment part subsequently.
Accompanying drawing explanation
Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for instructions, is used from explanation the present invention, but is not construed as limiting the invention with embodiment one below.In the accompanying drawings:
Fig. 1 is the schematic flow sheet of sound trade-mark searching method provided by the invention.
Embodiment
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in detail.Should be understood that, embodiment described herein, only for instruction and explanation of the present invention, is not limited to the present invention.
The invention provides a kind of sound trade-mark searching method, as shown in Figure 1, the method comprises the following steps: the sample trade mark in trade mark storehouse is divided into plural group according to duration; Obtain trade mark to be measured, and determine the group belonging to described trade mark to be measured according to the duration of trade mark to be measured; The similarity retrieval of trade mark to be measured is carried out: multiple tonic trains trade mark to be measured and sample trade mark being divided into equal length in determined group; Determine the audio frame sum P needed in trade mark to be measured and sample trade mark in the two-way audio sequence of comparison; Calculate the energy value needing each audio frame in the two-way audio sequence of comparison, with the energy ratio J of the audio frame of each correspondence in the two-way audio sequence obtaining needing comparison, judge to need two audio frames corresponding in the two-way audio sequence of comparison to be normal audio frame or anomalous audio frame according to the size of described energy ratio J, and calculate the average energy ratio K of the normal audio frame needed in the two-way audio sequence of comparison; Set the interval of described average energy ratio K, and determine to need the audio frame number Q of energy ratio within the interval of described average energy ratio K in the two-way audio sequence of comparison; For tonic train described in each road, calculate the quantity ratio of audio frame number Q and audio frame sum P, and judge the similarity of the two-way audio sequence needing comparison according to the ratio S needed between the quantity ratio of the two-way audio sequence of comparison; And the similarity of trade mark to be measured and sample trade mark is judged according to the similarity between the trade mark to be measured tonic train corresponding with sample trade mark.
According to technical scheme of the present invention, first divide into groups to the sample trade mark in trade mark storehouse, the sample trade mark in trade mark storehouse is divided into plural group by the duration according to sample trade mark.
For example, for comparatively very brief sample trade mark, such as blow a whistle sound, the tinkle of bells etc., the duration of trade mark is shorter, and correspondingly, such as, for comparatively long sample trade mark, one section of music etc., the duration of trade mark is longer.In practice, threshold interval can be pre-set for the grouping of sample trade mark in trade mark storehouse, namely threshold interval is set with the duration of whole trade mark, to be included into one by one in corresponding group by sample trade mark.It should be noted that, by presetting multiple threshold interval, the sample trade mark in trade mark storehouse can be divided into corresponding multiple group according to the needs of practical operation, the present invention does not carry out any restriction to this.
According to technical scheme of the present invention, in the process of carrying out trade mark retrieval, obtain trade mark to be measured, and determine the group belonging to described trade mark to be measured according to the duration of trade mark to be measured, in determined group, carry out the similarity retrieval of trade mark to be measured thus.Because the duration according to trade mark to be measured is retrieved in the sample trade mark group that it belongs to, the duration difference little (in determined threshold interval) of sample trade mark right by comparison, avoid and trade mark to be measured and the larger not approximate sample trade mark of difference are compared, without the need to retrieving in whole trade mark storehouse, shorten retrieval time, substantially increase trade mark recall precision.
For the similarity retrieval step of trade mark to be measured, technical scheme of the present invention is by being divided into minimum comparison unit tonic train by the melody of sample trade mark and trade mark to be measured, and to form tonic train audio frame one by one energy comparison value (i.e. amplitude) whether similarly judge between the tonic train that trade mark to be measured is corresponding with sample trade mark, and then judge the similarity of trade mark to be measured and sample trade mark, and technical scheme of the present invention eliminated anomalous audio frame before audio frame comparison, only keep the comparison of melody, avoid the interference of other sound in trade mark, can accurately identify audio frequency similarity, ensure that the accuracy that sound trade mark is retrieved.Below the similarity retrieval step of trade mark to be measured provided by the invention is specifically described.
Tonic train is the least unit of comparison, and audio frame is the least unit calculating tonic train energy value.According to technical scheme of the present invention, first trade mark to be measured and sample trade mark are divided into multiple tonic trains of equal length.
Audio frame is the set (value of usual N is 512 or 1024, and it is about 50 ~ 100ms that the time contained is about immediately) of N number of sampled point selected on audio frequency.In order to avoid the change of adjacent two audio frames is excessive, comprise one section of overlapping region between adjacent two audio frames in described tonic train, this overlapping region comprises M sampled point.Preferably, the value of described M is at least 1/2 of N value.The sampling frequency of the audio frequency that speech recognition is used is generally 8KHz or 16KHz.For 8KHz sampling frequency, if the length of an audio frame is 512 sampled points, then corresponding duration is 512/8000*1000=64ms.
According to audio frame determination tonic train.According to technical scheme of the present invention, a tonic train such as can get 50 ~ 100 audio frames, if each tonic train comprises 100 audio frames, then corresponding tonic train length is 64*100=6400ms.
Secondly, the audio frame sum P needed in trade mark to be measured and sample trade mark in the two-way audio sequence of comparison is determined.
Next, calculate the energy value needing each audio frame in the two-way audio sequence of comparison, with the energy ratio J of the audio frame of each correspondence in the two-way audio sequence obtaining needing comparison, judge to need two audio frames corresponding in the two-way audio sequence of comparison to be normal audio frame or anomalous audio frame according to the size of described energy ratio J, and calculate the average energy ratio K of the normal audio frame needed in the two-way audio sequence of comparison.
For the calculating of audio frame energy value, according to the characteristic of audio power symmetry, for audio frame only need calculate energy value be on the occasion of sampled point, the energy value of the sampled point comprised by audio frame adds up and just obtains the energy value of this audio frame.Therefore, according to technical scheme of the present invention, the account form of the energy value of each audio frame described is: by energy value in audio frame be on the occasion of sampled point energy value superposition, obtain the energy value of this audio frame.
In order to remove anomalous audio frame before audio frame comparison, ensureing the accuracy of audio frequency comparison, first needing the audio frame to needing in the two-way audio sequence of comparison to classify, calculate the energy ratio J of the audio frame of each correspondence in two-way audio sequence.Can set energy ratio threshold value t in advance, if need the energy ratio J of audio frame corresponding in the two-way audio sequence of comparison to be more than or equal to energy ratio threshold value t, then judge that two audio frames of this correspondence are anomalous audio frame; If need the energy ratio J of audio frame corresponding in the two-way audio sequence of comparison to be less than energy ratio threshold value t, then judge that two audio frames of this correspondence are normal audio frame.The value of described energy ratio threshold value t can set according to the needs of practical operation, such as, the span of described energy ratio threshold value t can be set to 8-12.Preferably, the value of described energy ratio threshold value t is 10.
For two that need to be confirmed as anomalous audio frame in the two-way audio sequence of comparison corresponding audio frames, no longer the energy value of described anomalous audio frame is calculated in the calculating of the average energy ratio K of follow-up normal audio frame, the impact of large energy ratio J on the average energy ratio K of normal audio frame can be eliminated.Thus, in the calculating of follow-up average energy ratio K, only calculate the average energy ratio K of the normal audio frame needed in the two-way audio sequence of comparison.
According to technical scheme of the present invention, described need the account form of the average energy ratio K of normal audio frame in the two-way audio sequence of comparison for: the energy value of the normal audio frame in tonic train described in each road is averaging respectively, to obtain the average energy of normal audio frame in tonic train described in each road, using the ratio of the average energy of normal audio frame in the two-way audio sequence needing comparison as average energy ratio K.
Next, set the interval of described average energy ratio K, and determine to need the audio frame number Q of energy ratio within the interval of described average energy ratio K in the two-way audio sequence of comparison.According to technical scheme of the present invention, the interval of described average energy ratio K can be set as (K-r, K+r), wherein the span of r can set according to the needs of practical operation, and the span of such as r can be set to 1-3.Preferably, the value of r is 2, and namely the interval of described average energy ratio K is (K-2, K+2).According to the interval of set described average energy ratio K, determine to need the audio frame number Q of energy ratio within the interval of described average energy ratio K in the two-way audio sequence of comparison.Wherein, the audio frame number Q of energy ratio within the interval of described average energy ratio K can comprise anomalous audio frame.
Next, enter the similarity determining step of the two-way audio sequence needing comparison: for tonic train described in each road, calculate the quantity ratio of audio frame number Q and audio frame sum P, and judge the similarity of the two-way audio sequence needing comparison according to the ratio S needed between the quantity ratio of the two-way audio sequence of comparison.
In general, if two-way audio content is identical, then the energy ratio of identical, the audio frame of the overall trend of the two can fluctuate in a less interval; Otherwise if two-way audio content is not identical, then the energy ratio of audio frame does not have rule, and there will be the larger situation of energy ratio.Therefore, method provided by the invention is first by removing the anomalous audio frame needed in the two-way audio sequence of comparison, obtain the average energy ratio K of normal audio frame, then judged the similarity of two-way audio sequence by the audio frame number Q in the interval of the average energy ratio K of statistics normal audio frame.
According to technical scheme of the present invention, if need the ratio S between the quantity ratio of the two-way audio sequence of comparison to be less than threshold value w, then judge the two-way audio sequence similarity needing comparison; If need the ratio S between the quantity ratio of the two-way audio sequence of comparison to be more than or equal to threshold value w, then judge to need the two-way audio sequence of comparison dissimilar.
The value of described threshold value w can set according to the needs of practical operation, such as, the span of described threshold value w can be set to 0.1-0.2.Preferably, the value of described threshold value w can be set to 0.15.That is, if need the ratio S between the quantity ratio of the two-way audio sequence of comparison to be less than 0.15, then judge the two-way audio sequence similarity needing comparison, otherwise judge that the two-way audio sequence needing comparison is dissimilar.
According to above-mentioned steps, user can adopt the criterion of sound trade mark, based on the similarity between the tonic train that trade mark to be measured is corresponding with sample trade mark, thus judges the similarity of trade mark to be measured and sample trade mark.
According to technical scheme of the present invention, preferably, described sound trade-mark searching method is further comprising the steps of: judging in all dissimilar situation of each sample trade mark in the group belonging to trade mark to be measured and this trade mark to be measured, carry out the similarity retrieval of this trade mark to be measured in group outside the group belonging to this trade mark to be measured, the accuracy of similarity retrieval can be ensured thus better.
Sound trade-mark searching method provided by the invention carries out the similarity retrieval of trade mark to be measured in the determined group of trade mark to be measured, avoid and trade mark to be measured and the larger not approximate sample trade mark of difference are compared, without the need to retrieving in whole trade mark storehouse, shorten retrieval time, substantially increase trade mark recall precision, and can accurately identify audio frequency similarity, ensure that the accuracy that sound trade mark is retrieved.
Below the preferred embodiment of the present invention is described in detail by reference to the accompanying drawings; but; the present invention is not limited to the detail in above-mentioned embodiment; within the scope of technical conceive of the present invention; can carry out multiple simple variant to technical scheme of the present invention, these simple variant all belong to protection scope of the present invention.
It should be noted that in addition, each the concrete technical characteristic described in above-mentioned embodiment, in reconcilable situation, can be combined by any suitable mode.
In addition, also can carry out combination in any between various different embodiment of the present invention, as long as it is without prejudice to thought of the present invention, it should be considered as content disclosed in this invention equally.
Claims (13)
1. a sound trade-mark searching method, is characterized in that, the method comprises the following steps:
Sample trade mark in trade mark storehouse is divided into plural group according to duration;
Obtain trade mark to be measured, and determine the group belonging to described trade mark to be measured according to the duration of trade mark to be measured;
The similarity retrieval of trade mark to be measured is carried out in determined group:
Trade mark to be measured and sample trade mark are divided into multiple tonic trains of equal length;
Determine the audio frame sum P needed in trade mark to be measured and sample trade mark in the two-way audio sequence of comparison;
Calculate the energy value needing each audio frame in the two-way audio sequence of comparison, with the energy ratio J of the audio frame of each correspondence in the two-way audio sequence obtaining needing comparison, judge to need two audio frames corresponding in the two-way audio sequence of comparison to be normal audio frame or anomalous audio frame according to the size of described energy ratio J, and calculate the average energy ratio K of the normal audio frame needed in the two-way audio sequence of comparison;
Set the interval of described average energy ratio K, and determine to need the audio frame number Q of energy ratio within the interval of described average energy ratio K in the two-way audio sequence of comparison;
For tonic train described in each road, calculate the quantity ratio of audio frame number Q and audio frame sum P, and judge the similarity of the two-way audio sequence needing comparison according to the ratio S needed between the quantity ratio of the two-way audio sequence of comparison; And
The similarity of trade mark to be measured and sample trade mark is judged according to the similarity between the tonic train that trade mark to be measured is corresponding with sample trade mark.
2. method according to claim 1, it is characterized in that, described need the account form of the average energy ratio K of normal audio frame in the two-way audio sequence of comparison for: the energy value of the normal audio frame in tonic train described in each road is averaging respectively, to obtain the average energy of normal audio frame in tonic train described in each road, using the ratio of the average energy of normal audio frame in the two-way audio sequence needing comparison as average energy ratio K.
3. method according to claim 1, is characterized in that, the interval of described average energy ratio K is (K-r, K+r), and wherein the span of r is 1-3.
4. method according to claim 1, is characterized in that, described audio frame is the set of N number of sampled point selected on audio frequency.
5. method according to claim 4, is characterized in that, the account form of the energy value of each audio frame described is: by energy value in audio frame be on the occasion of sampled point energy value superposition, obtain the energy value of this audio frame.
6. method according to claim 4, is characterized in that, comprise one section of overlapping region between adjacent two audio frames in described tonic train, this overlapping region comprises M sampled point.
7. method according to claim 6, is characterized in that, the value of described M is at least 1/2 of N value.
8. method according to claim 1, it is characterized in that, if need the energy ratio J of audio frame corresponding in the two-way audio sequence of comparison to be more than or equal to energy ratio threshold value t, two audio frames then judging this correspondence are anomalous audio frame, if need the energy ratio J of audio frame corresponding in the two-way audio sequence of comparison to be less than energy ratio threshold value t, then judge that two audio frames of this correspondence are normal audio frame.
9. method according to claim 8, is characterized in that, the span of described energy ratio threshold value t is 8-12.
10. method according to claim 1, is characterized in that, if need the ratio S between the quantity ratio of the two-way audio sequence of comparison to be less than threshold value w, then judges the two-way audio sequence similarity needing comparison.
11. methods according to claim 1, is characterized in that, if need the ratio S between the quantity ratio of the two-way audio sequence of comparison to be more than or equal to threshold value w, then judge to need the two-way audio sequence of comparison dissimilar.
12. methods according to claim 10 or 11, it is characterized in that, the span of described threshold value w is 0.1-0.2.
13. methods according to claim 1, is characterized in that, described method is further comprising the steps of:
Judging, in all dissimilar situation of each sample trade mark in the group belonging to trade mark to be measured and this trade mark to be measured, in the group outside the group belonging to this trade mark to be measured, to carry out the similarity retrieval of this trade mark to be measured.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410752310.9A CN104464754A (en) | 2014-12-11 | 2014-12-11 | Sound brand search method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410752310.9A CN104464754A (en) | 2014-12-11 | 2014-12-11 | Sound brand search method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104464754A true CN104464754A (en) | 2015-03-25 |
Family
ID=52910698
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410752310.9A Pending CN104464754A (en) | 2014-12-11 | 2014-12-11 | Sound brand search method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104464754A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108985584A (en) * | 2018-06-27 | 2018-12-11 | 广州朝舜网络科技有限公司 | A kind of trade mark intelligent analysis method, device, terminal and storage medium |
| CN111681671A (en) * | 2020-05-20 | 2020-09-18 | 浙江大华技术股份有限公司 | Abnormal sound identification method and device and computer storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1629901A (en) * | 2003-12-15 | 2005-06-22 | 联发科技股份有限公司 | Kara-OK score device and method |
| US7035775B2 (en) * | 2002-07-09 | 2006-04-25 | Sony Corporation | Similar time series detection method and apparatus, program and recording medium |
| CN101021854A (en) * | 2006-10-11 | 2007-08-22 | 鲍东山 | Audio analysis system based on content |
| CN101593517A (en) * | 2009-06-29 | 2009-12-02 | 北京市博汇科技有限公司 | A kind of audio frequency comparison system and audio power comparison method thereof |
| CN102073636A (en) * | 2009-10-30 | 2011-05-25 | 索尼株式会社 | Program climax search method and system |
| CN102842310A (en) * | 2012-08-10 | 2012-12-26 | 上海协言科学技术服务有限公司 | Method for extracting and utilizing audio features for repairing Chinese national folk music audios |
| CN102956237A (en) * | 2011-08-19 | 2013-03-06 | 杜比实验室特许公司 | Method and device for measuring content consistency and method and device for measuring similarity |
| US20130138232A1 (en) * | 2011-11-28 | 2013-05-30 | Electronics And Telecommunications Research Institute | Apparatus and method for extracting highlight section of music |
| CN104091598A (en) * | 2013-04-18 | 2014-10-08 | 腾讯科技（深圳）有限公司 | Audio file similarity calculation method and device |
| CN104157296A (en) * | 2014-07-28 | 2014-11-19 | 腾讯科技（深圳）有限公司 | Audio frequency evaluative method and device |
-
2014
- 2014-12-11 CN CN201410752310.9A patent/CN104464754A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7035775B2 (en) * | 2002-07-09 | 2006-04-25 | Sony Corporation | Similar time series detection method and apparatus, program and recording medium |
| CN1629901A (en) * | 2003-12-15 | 2005-06-22 | 联发科技股份有限公司 | Kara-OK score device and method |
| CN101021854A (en) * | 2006-10-11 | 2007-08-22 | 鲍东山 | Audio analysis system based on content |
| CN101593517A (en) * | 2009-06-29 | 2009-12-02 | 北京市博汇科技有限公司 | A kind of audio frequency comparison system and audio power comparison method thereof |
| CN102073636A (en) * | 2009-10-30 | 2011-05-25 | 索尼株式会社 | Program climax search method and system |
| CN102956237A (en) * | 2011-08-19 | 2013-03-06 | 杜比实验室特许公司 | Method and device for measuring content consistency and method and device for measuring similarity |
| US20130138232A1 (en) * | 2011-11-28 | 2013-05-30 | Electronics And Telecommunications Research Institute | Apparatus and method for extracting highlight section of music |
| CN102842310A (en) * | 2012-08-10 | 2012-12-26 | 上海协言科学技术服务有限公司 | Method for extracting and utilizing audio features for repairing Chinese national folk music audios |
| CN104091598A (en) * | 2013-04-18 | 2014-10-08 | 腾讯科技（深圳）有限公司 | Audio file similarity calculation method and device |
| CN104157296A (en) * | 2014-07-28 | 2014-11-19 | 腾讯科技（深圳）有限公司 | Audio frequency evaluative method and device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108985584A (en) * | 2018-06-27 | 2018-12-11 | 广州朝舜网络科技有限公司 | A kind of trade mark intelligent analysis method, device, terminal and storage medium |
| CN111681671A (en) * | 2020-05-20 | 2020-09-18 | 浙江大华技术股份有限公司 | Abnormal sound identification method and device and computer storage medium |
| CN111681671B (en) * | 2020-05-20 | 2023-03-10 | 浙江大华技术股份有限公司 | Abnormal sound identification method and device and computer storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103500579B (en) | Audio recognition method, Apparatus and system | |
| US20220060587A1 (en) | Methods and apparatus to identify media using hybrid hash keys | |
| KR102614021B1 (en) | Audio content recognition method and device | |
| EP3163466B1 (en) | Trajectory data inquiry method and apparatus | |
| US10368127B2 (en) | Methods and apparatus to identify and credit media using ratios of media characteristics | |
| CN101821734A (en) | Detection and classification of matches between time-based media | |
| US20140351273A1 (en) | System and method for searching information | |
| CN103093761A (en) | Audio fingerprint retrieval method and retrieval device | |
| CN104318327A (en) | Predictive parsing method for track of vehicle | |
| CN115908080B (en) | Carbon emission optimization method and system based on multidimensional data analysis | |
| RU2008140138A (en) | PRODUCT DATA FOR SERVICES | |
| GB2525365A (en) | A system and methods thereof for consumer purchase identification for value-added tax (VAT) reclaim | |
| CN108255840A (en) | A kind of recommendation method and system of song | |
| ATE363779T1 (en) | PRELIMINARY TESTING AND CERTIFICATION OF MULTIPLE ACCESS CODES | |
| CN109712642B (en) | Accurate and rapid advertisement broadcasting monitoring method | |
| CN104464754A (en) | Sound brand search method | |
| US20180191850A1 (en) | Methods and apparatus to facilitate meter to meter matching for media identification | |
| CN112468754A (en) | Method and device for acquiring stroke record data based on audio and video recognition technology | |
| KR20180073299A (en) | Apparatus and method for detecting abnormal event using statistics | |
| WO2014020449A2 (en) | Identifying audio stream content | |
| WO2012037315A3 (en) | Customer focused keyword search in an enterprise | |
| JP6110281B2 (en) | Moving unit prediction model generation apparatus, moving unit prediction model generation method, and moving unit prediction model generation program | |
| Pinelli et al. | A methodology for denoising and generating bus infrastructure data | |
| WO2005109253A3 (en) | Method for the computer-assisted evaluation of the prognosis of characteristic values of a technical system, performed by means of a predictive model | |
| CN103354097A (en) | Video data storing method and system based on disk |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 102455, Beijing, Changyang Town, Changyang, Fangshan District village, Nanjing railway, 110 meters East northbound Applicant after: CIPRUN MOBILE INTERCONNECTION SCIENCE & TECHNOLOGY CO., LTD. Address before: 102455, Beijing, Changyang Town, Changyang, Fangshan District village, Nanjing railway, 110 meters East northbound Applicant before: BEIJING CIPRUN MOBILE INTERNET TECHNOLOGY CO., LTD. |
|
| COR | Change of bibliographic data | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150325 |