CN101827256A - Voiceprint authentication and digital watermark-based remote digitized court hearing method - Google Patents
Voiceprint authentication and digital watermark-based remote digitized court hearing method Download PDFInfo
- Publication number
- CN101827256A CN101827256A CN200910079537A CN200910079537A CN101827256A CN 101827256 A CN101827256 A CN 101827256A CN 200910079537 A CN200910079537 A CN 200910079537A CN 200910079537 A CN200910079537 A CN 200910079537A CN 101827256 A CN101827256 A CN 101827256A
- Authority
- CN
- China
- Prior art keywords
- participant
- proceedings
- video
- court
- 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
Images
Landscapes
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The invention discloses a voiceprint authentication and digital watermark-based remote digitized court hearing method and relates to the technical fields of remote identity authentication, remote video and audio transmission and synchronism and the like. In a conventional mode, every part of litigant participants is required to attend the court hearing in an adjudication division provided by a court, so the litigation cost is high, and the efficiency is low. The method comprises the following steps of: performing remote authentication on the identity of every part of the litigant participants through voiceprint authentication; acquiring video and audio data of every part of the litigant participants and embedding the voiceprint characteristics of the litigant participants which serve as the digital watermark into the video and audio data; simultaneously, sending the video and audio data of every part of litigant participants to the other part of the litigant participants, extracting and verifying the digital watermark in the video and audio data by the other litigant participants, and synchronizing the video and audio data after the verification; and finally displaying the video and audio data of every part of the litigant participants on display equipment. Due to the adoption of the method, remote digitized court hearing is realized; and the method has the advantages of low litigation cost, high inquisition efficiency and high safety.
Description
Technical field
The present invention relates to long-distance identity-certifying, long-range audio transmission and the technical field such as synchronous of looking, be specifically related to a kind of remote digital court's trial method, be applicable to that law court carries out in the process of court's trial case based on voiceprint and digital watermarking.
Background technology
Regulation according to existing law, case mainly classify civil case, administrative case and criminal case.Accordingly, the trial of case just is divided into the trial of civil case, the trial of administrative case and the trial of criminal case.The difference of the difference of case type and case concrete condition can make that the participant in proceedings in the court trial process is different.For example, among the civil case participant in proceedings judge, plaintiff, defendant must be arranged, also have the 3rd people, witness, surveyor etc.; The participant in proceedings of criminal case also must have the public prosecutor.The court's trial program can roughly be summarized as follows several stages: participant in proceedings's identity is carried out the stage of recognition (this stage carries out) before court's trial; The court investigation stage (plaintiff and defendant carry out case statement and proof respectively in this stage, and the evidence that the other side is provided carries out cross-examination, if the witness is arranged, then also needs the attendance of witness to testify); The court debate stage (central issue being had argument) by plaintiff and defendant; The closing statement stage (plaintiff and defendant carry out closing statement respectively).
Along with development of science and technology, law courts at different levels begin tentative with technological means support court's trial work gradually.For example, adopt multimedia equipment to assist court's trial, court trial process is carried out network direct broadcasting etc.This application is in the level of manipulating of equipment basically, does not fundamentally change traditional court's trial mode, and the process of court's trial still needs participant in proceedings each side to carry out in the adjudication division that law court provides.On the one hand, the lawsuit cost is higher, because law court need provide the adjudication division, other participants in proceedings need transport cost etc.; On the other hand, also can influence the trial quality of case in some cases, for example the witness is because certain is former thereby can not testify or fear to retaliate and be reluctant to testify etc.In addition, owing to the finiteness of law court's adjudication division's resource and party because of reasonable ground reason such as can't present in court, cause the delay of the time limit for handling the case, influenced the trial efficient of case.
Summary of the invention
At the defective that exists in the prior art, the purpose of this invention is to provide a kind of based on voiceprint and digital watermarking, the lawsuit cost is low, trial efficient height, remote digital court's trial method that fail safe is good.
To achieve these goals, the technical solution used in the present invention is as follows: a kind of remote digital court's trial method based on voiceprint and digital watermarking may further comprise the steps:
Before court's trial, set up all participant in proceedingss' sound-groove model;
In court trial process, the sound of gathering the participant in proceedings also utilizes sound-groove model that the identity of participant in proceedings each side is carried out remote authentication, and described participant in proceedings each side is in different places, interconnects by network;
Gather the video/audio of participant in proceedings each side, and be embedded into this participant in proceedings's vocal print feature in the video/audio of this participant in proceedings side as digital watermarking;
The video/audio of each participant in proceedings side is sent to other participant in proceedings sides simultaneously, the digital watermarking that other participant in proceedings sides extract in the video/audio is verified, checking by after video/audio is carried out synchronously, at last the video/audio of each participant in proceedings side is shown on display device simultaneously.
Aforesaid a kind of remote digital court's trial method based on voiceprint and digital watermarking, the method of the described participant in proceedings's of foundation sound-groove model is: at first set up general background sound-groove model by sample vocal print feature, extract participant in proceedings's vocal print feature then, according to this participant in proceedings's vocal print feature modification background sound-groove model, with the sound-groove model of amended background sound-groove model as this participant in proceedings.
Aforesaid a kind of remote digital court's trial method based on voiceprint and digital watermarking, described collection participant in proceedings's sound and the process of utilizing sound-groove model that the identity of participant in proceedings each side is carried out remote authentication may further comprise the steps:
Gather participant in proceedings's voice, extract the efficient voice section;
From the efficient voice section, extract the vocal print feature;
The sound-groove model of vocal print feature with this participant in proceedings who sets up in advance mated.
Aforesaid a kind of remote digital court's trial method based on voiceprint and digital watermarking, when the participant in proceedings formally makes a speech, just to its authentication once by vocal print, if, then send warning message to all participant in proceedings sides not by checking.
Aforesaid a kind of remote digital court's trial method based on voiceprint and digital watermarking, the process of the video/audio of described collection participant in proceedings each side may further comprise the steps:
Video data and voice data are gathered respectively;
Video data and the voice data gathered are carried out buffer memory;
When the video data of buffer memory and voice data reach predefined size, video data and voice data are encoded respectively.
Aforesaid a kind of remote digital court's trial method based on voiceprint and digital watermarking is embedded into participant in proceedings's vocal print feature in the original video and audio code stream before the coding; Perhaps in video and audio data coding process, the vocal print feature is embedded in video and the voice data as digital watermarking; Perhaps the vocal print feature is embedded in the video and audio code stream behind the coding as digital watermarking.
Aforesaid a kind of remote digital court's trial method based on voiceprint and digital watermarking, this method also comprises the step that decoded video/audio is handled, and described decoded video/audio is handled comprises and strengthen video definition, audio defeat and personage's head portrait in the video is made mosaic processing.
Aforesaid a kind of remote digital court's trial method based on voiceprint and digital watermarking is divided into a plurality of viewing areas with display device, and each viewing area is used to show participant in proceedings one side's the audio frequency of looking.When a certain viewing area did not have displaying contents, then corresponding with the participant in proceedings side of speech viewing area was by corresponding amplification; After this participant in proceedings holds speech to finish, narrow down to original size automatically.
Adopt the method for the invention, participant in proceedings each side can be in different geographical position in the court trial process, the unified trial court that provides to law court is provided engages in a lawsuit, thereby saved the lawsuit cost of each side.Because court trial process is not subjected to the restriction of region, phenomenon is incured loss through delay in the lawsuit of having avoided causing because of problems such as traffic, has shortened the time limit that law court hears a case, and has improved trial efficient.In addition, confirm participant in proceedings's identity by voiceprint, and with its as digital watermarking be embedded into look, in the voice data, authenticity that not only can actio finium regundorum participant identity, need not to increase extra authenticating device, and can guarantee the fail safe of video/audio when Network Transmission.
Description of drawings
Fig. 1 is the method for the invention flow chart;
Fig. 2 is the voiceprint flow chart;
Fig. 3 is a display device subregion schematic diagram.
Embodiment
Describe the present invention below in conjunction with embodiment and accompanying drawing.
Core concept of the present invention is: in the court trial process, participant in proceedings each side can be in different places, interconnect by network, utilize and to look the audio frequency of looking that audio collecting device is gathered each side, with looking on the display device that audio frequency is simultaneously displayed on each side of each side.Authenticity and the fail safe of video/audio in network transmission process in order to ensure participant in proceedings's identity, adopt the mode of voiceprint and the vocal print feature is embedded in the video/audio as watermark, prevent that video/audio from being distorted in network transmission process.
Vocal print is the sound wave spectrum that carries language message that the electricity consumption acoustic instrument shows.Emergence of human language is complex physical physical process between human body speech center and the vocal organs, the generation organ (tongue, tooth, larynx, lung, nasal cavity) that the people used in when speech is everyone widely different aspect size and the form, so any two people's vocal print collection of illustrative plates is all variant.But everyone existing relative stability of voice acoustic feature has variability again, is not absolute, unalterable.However, because everyone vocal organs all are not quite similar, therefore in the ordinary course of things, still can distinguish different people's sound or judge whether is same people's sound.
In the court trial process, adopt speech recognition civil litigant participants identity to need not to increase extra hardware device, and can when the participant in proceedings makes a speech, collect voice at any time and verify, need not participant in proceedings's cooperation, so the present invention adopts the voiceprint mode to come the authenticity of actio finium regundorum participant identity.
Fig. 1 has shown the method for the invention flow process, mainly may further comprise the steps.
(1) before court's trial, sets up all participant in proceedingss' sound-groove model.
Before court's trial, at first need to confirm the identity of participant in proceedings each side.The participant in proceedings is different and different according to concrete case type and case characteristics, comprises judge, plaintiff, defendant, public prosecutor, the 3rd people, witness etc.Participant in proceedings each side can be in different places during court's trial, carries out long-distance identity-certifying by network.
After having determined the participant in proceedings, need before court's trial, set up all participant in proceedingss' sound-groove model, deposit in the database, the benchmark model of checking participant in proceedings identity during as court's trial.
Application on Voiceprint Recognition can be divided into text relevant with two kinds of text-independent.The Application on Voiceprint Recognition that text is relevant needs identified person's content pronunciation according to the rules, accurately set up sound-groove model, and when identification also must be according to the rules the content pronunciation, therefore can reach recognition effect preferably, but need identified person's cooperation, if the content of identified person's pronunciation and regulation does not meet, then can't correctly discern.Then do not stipulate speaker's pronunciation content with the identification of text-independent, this RM is set up comparatively difficulty of model, but need not identified person's cooperation.
Because present embodiment when the participant in proceedings formally makes a speech, is just collected this participant in proceedings's voice and verified, can guarantee the authenticity of participant in proceedings's identity in the whole court trial process like this, therefore adopt RM with text-independent.
When setting up model, at first collect people's abundant under the particular channel sample voice, for example, collect each 100 people of men and women, everyone is no less than 1 minute sample voice, is used for the proprietary average vocal print characteristic distribution of approximate description.After then these voice being extracted the vocal print characteristic parameter, put together and carry out the iteration cluster, obtain several Gaussian Mixture, as general vocal print background model (UBM).
Need voice are carried out preliminary treatment before extracting speech characteristic parameter, because in the raw tone except containing the significant content that speaker self sends, pause quiet in addition or meaningless part such as background noise interference, so should judge the validity of each voice segments earlier, get rid of insignificant part, only from the efficient voice section, extract the vocal print feature.Extract the vocal print feature and can adopt method as described below: the efficient voice sampled point is divided into one group of fenestella (promptly dividing frame) with certain-length, has local overlapping between adjacent windows.Each frame inside at first applies hamming (Hamming) window function, and carries out high boost, does fast Fourier transform (FFT) then, obtains frequency spectrum.Again frequency spectrum is applied by Mel scale rule bank of filters that arrange, that contain some triangle filters, the output valve of each filter of obtaining is carried out discrete cosine transform (DCT), obtain one group of coefficient, be Mel frequency cepstrum parameter (MFCC) vector, then the MFCC vector of consecutive frame is asked poor by dimension, and the difference vector that obtains is spliced on the former MFCC vector, the parameter that obtains is as the vocal print characteristic sequence.
After having set up the vocal print background model, gather civil litigant participants's sufficiently long efficient voice, for example gather 1 minute efficient voice, utilize said method to extract the vocal print feature, again according to the adaptive algorithm of maximum a posteriori probability (MAP), with the vocal print feature correction UBM that extracts, thereby (everyone vocal print feature can be regarded as a kind of probability distribution at high-dimensional feature space to obtain a new gauss hybrid models, can come match with a plurality of Gaussian Profile, this class set is collectively referred to as makes gauss hybrid models), as this participant in proceedings's sound-groove model, deposit in the database coupling benchmark in by this civil litigant participants of speech recognition during as court's trial.
Before court's trial, set up all participant in proceedingss' sound-groove model by said method.
Fig. 2 has shown the method flow of voiceprint, may further comprise the steps: at first gather participant in proceedings's voice, extract the efficient voice section; From the efficient voice section, extract the vocal print feature then; The vocal print feature of extracting is sent to the server of storing sound-groove model by network, on server, the sound-groove model of vocal print feature with this participant in proceedings who sets up in advance mated at last.
Present embodiment is in court trial process, when the participant in proceedings formally makes a speech, just by vocal print to its authentication once.The process of checking is carried out on the backstage, if authentication failed then sends warning message to participant in proceedings each side.
(2) gather the video/audio of participant in proceedings each side, and this participant in proceedings's vocal print feature is embedded in the video/audio of this participant in proceedings place side as digital watermarking.
Participant in proceedings each side is equipped with video capture device and audio collecting device, and video and audio frequency adopt independent passage to gather.At first video data and the voice data that collects carried out buffer memory, when the size of data in buffer reaches preset value, video data and voice data are encoded, video data and voice data after will encoding then are distributed to participant in proceedings each side.In video data and voice data, all comprise time point information, the reference when looking audio sync.
Need to prove: when video and audio collection, have only when participant in proceedings's one upright formula is made a speech, just need to gather simultaneously speech one side's video and audio frequency, its other party is then only gathered video information and is got final product, and need not to gather audio-frequency information.
In said process, the vocal print feature of extracting during with authentication is embedded into respectively and looks in sound data and the voice data as watermark.Digital watermarking can embed in original video and voice data code stream, also can embed in coding stage, can also embed behind coding.Watermark directly is embedded in the data bit-rate that not only can increase video and audio stream in original video and the audio stream, and watermark goes out the active phenomenon easily behind the coding, also can reduce the quality of video and audio frequency.Watermarked in cataloged procedure, generally can not increase the data bit-rate of video and audio stream, but can reduce the quality of video and audio frequency.Watermarked behind coding, computation complexity is low on the one hand, can not cause the decline of video and audio quality; Define the size of watermarked data volume on the other hand owing to the restriction of compression bit rate.Comprehensive above reason and in conjunction with the characteristics of court's trial, the present embodiment employing is watermarked in cataloged procedure, because the quality requirement to video and audio frequency is not high in the court trial process, but the efficient of the network delivery of video and voice data is had relatively high expectations.
(3) video/audio with each participant in proceedings side sends to other participant in proceedings sides simultaneously, the digital watermarking that other participant in proceedings sides extract in the video/audio is verified, checking by after video/audio is carried out synchronously, at last the video/audio of each participant in proceedings side is shown on display device simultaneously.
When participant in proceedings each side receive its other party look audio frequency the time, the decoding process in extract digital watermarking simultaneously, mate with original vocal print feature then as digital watermarking.If coupling is coincide, prove that then this section looks audio frequency and do not distorted in network transmission process.
After the video/audio decoding, can handle again, for example strengthen video definition, audio defeat and personage's head portrait in the video is made mosaic processing etc. video/audio.In court trial process, if the witness does not wish the openly appearance feature of oneself, then its face can be carried out mosaic processing, can play protective effect like this to the witness.
Next video data and voice data are carried out Synchronous Processing, only need the side's that makes a speech video and audio frequency are carried out synchronously herein, what its other party was only gathered is video data, therefore need not synchronous operation.When participant in proceedings side (except that we, be video and audio frequency place side) the looking of same time period, audio frequency are all finished synchronously after, the unified transmission to all participant in proceedings sides ordered, and plays this section simultaneously and look audio frequency on the display device of all participant in proceedings sides.Like this, can guarantee content displayed unanimity on the display device of all participant in proceedings sides.
In participant in proceedings each side many display devices can be set, be used for showing respectively the audio frequency of looking of participant in proceedings each side; Also a display device can be carried out subregion, every district shows participant in proceedings one side's the audio frequency of looking.As shown in Figure 3, display device is divided into judge viewing area, plaintiff viewing area, defendant viewing area, witness viewing area, evidence viewing area, warning message viewing area.When a certain district did not have displaying contents, as witness viewing area, warning message viewing area, the size of other viewing areas was by corresponding amplification.In the present embodiment, when a certain participant in proceedings speech, and a certain viewing area is not when having displaying contents, and then corresponding with participant in proceedings viewing area is amplified accordingly; After this participant in proceedings makes a speech end, then narrow down to original size automatically.
In addition,, need plaintiff and defendant's twocouese court to lodge a proof,, also need the witness that testimony is provided if the witness is arranged in the court investigation stage of court's trial.In the court's trial that has the witness to participate in, also need to be provided with witness one side, when the witness testify finish after, the witness is convenient to be cut off with other participant in proceedings sides.As the plaintiff or defendant lodges a proof and the practical work that this evidence will prove is carried out narrative tense, demonstration and plaintiff or defendant are stating corresponding digitlization evidence in the evidence viewing area of the display device of each side.Digitlization evidence described herein is to make before court's trial according to the evidence that plaintiff and defendant in the proof deadline submit to.For example, material evidence can be taken into picture, and documented evidence, expert's conclusion, records of inquests etc. can be scanned into e-file.When evidence demonstration, the display device that can control participant in proceedings each side shows simultaneously.
After the witness provides testimony, after perhaps court's trial finishes, need confirm the court's trial notes of clerk's record.At this moment, need the participant in proceedings side of sign or seal that digital signature equipment or electronic seal equipment should be set.By sending to the participant in proceedings side that needs sign or seal, this participant in proceedings side utilizes digital signature equipment or electronic seal equipment to return after carrying out sign or seal on the e-file and gives method school then with the e-file that writes down in the court trial process in law court side.In order to guarantee that e-file is distorted in network transmission process, participant in proceedings's vocal print feature can be embedded in the e-file as digital watermarking equally.
In court trial process, the public can audit the case that can try in public, so generally all provides visitors seats in the adjudication division of law court, is convenient to the public using case trial electric process is exercised supervision, but the public does not have the right of speech in court trial process, must not violate court's discipline yet.In the case that can try in public, with in the looking audio frequency and send to its other party of participant in proceedings one side, send to the server end that is used for network direct broadcasting, server end carries out network direct broadcasting after it is handled.Also can be after court's trial finish, all of storage are looked audio file be integrated into one section complete audio file of looking, announce on the internet, be used for that the public is online to watch and download.Law court not only need not provide visitors seats like this, and has avoided the influence of auditor to the litigious party, and violates the influence of the auditor of court's discipline to the using case trial electric process.
In court trial process, also need all audio files of looking are backed up, so that law court side keeps counterfoil, consult the hearing process of this case after being convenient to, or exchange between the law court, learn to use, can also be published on the Internet after its processing, watch and download for the public.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these are revised and modification belongs within the scope of claim of the present invention and equivalent technology thereof, then the present invention also is intended to comprise these changes and modification interior.
Claims (9)
1. remote digital court's trial method based on voiceprint and digital watermarking may further comprise the steps:
Before court's trial, set up all participant in proceedingss' sound-groove model;
In court trial process, the sound of gathering the participant in proceedings also utilizes sound-groove model that the identity of participant in proceedings each side is carried out remote authentication, and described participant in proceedings each side is in different places, interconnects by network;
Gather the video/audio of participant in proceedings each side, and be embedded into this participant in proceedings's vocal print feature in the video/audio of this participant in proceedings side as digital watermarking;
The video/audio of each participant in proceedings side is sent to other participant in proceedings sides simultaneously, the digital watermarking that other participant in proceedings sides extract in the video/audio is verified, checking by after video/audio is carried out synchronously, at last the video/audio of each participant in proceedings side is shown on display device simultaneously.
2. a kind of remote digital court's trial method as claimed in claim 1 based on voiceprint and digital watermarking, it is characterized in that, the method of the described participant in proceedings's of foundation sound-groove model is: at first set up general background sound-groove model by sample vocal print feature, extract participant in proceedings's vocal print feature then, according to this participant in proceedings's vocal print feature modification background sound-groove model, with the sound-groove model of amended background sound-groove model as this participant in proceedings.
3. a kind of remote digital court's trial method as claimed in claim 2 based on voiceprint and digital watermarking, it is characterized in that described collection participant in proceedings's sound and the process of utilizing sound-groove model that the identity of participant in proceedings each side is carried out remote authentication may further comprise the steps:
Gather participant in proceedings's voice, extract the efficient voice section;
From the efficient voice section, extract the vocal print feature;
The sound-groove model of vocal print feature with this participant in proceedings who sets up in advance mated.
4. a kind of remote digital court's trial method as claimed in claim 3 based on voiceprint and digital watermarking, it is characterized in that: when the participant in proceedings formally makes a speech, just to its authentication once, if, then send warning message to all participant in proceedings sides not by checking by vocal print.
5. as the described a kind of remote digital court's trial method of one of claim 1 to 4, it is characterized in that described process of gathering the video/audio of participant in proceedings each side may further comprise the steps based on voiceprint and digital watermarking:
Video data and voice data are gathered respectively;
Video data and the voice data gathered are carried out buffer memory;
When the video data of buffer memory and voice data reach predefined size, video data and voice data are encoded respectively.
6. a kind of remote digital court's trial method based on voiceprint and digital watermarking as claimed in claim 5 is characterized in that: participant in proceedings's vocal print feature is embedded in the coding preceding original video and audio code stream; Perhaps in video and audio data coding process, the vocal print feature is embedded in video and the voice data as digital watermarking; Perhaps the vocal print feature is embedded in the video and audio code stream behind the coding as digital watermarking.
7. as the described a kind of remote digital court's trial method of one of claim 1 to 4 based on voiceprint and digital watermarking, it is characterized in that: described method also comprises the step that decoded video/audio is handled, and described decoded video/audio is handled comprises and strengthen video definition, audio defeat and personage's head portrait in the video is made mosaic processing.
8. as the described a kind of remote digital court's trial method based on voiceprint and digital watermarking of one of claim 1 to 4, it is characterized in that: display device is divided into a plurality of viewing areas, and each viewing area is used to show participant in proceedings one side's the audio frequency of looking.
9. a kind of remote digital court's trial method based on voiceprint and digital watermarking as claimed in claim 8 is characterized in that: when a certain viewing area of display device did not have displaying contents, then corresponding with the participant in proceedings side of speech viewing area was by corresponding amplification; After this participant in proceedings holds speech to finish, narrow down to original size automatically.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910079537A CN101827256A (en) | 2009-03-06 | 2009-03-06 | Voiceprint authentication and digital watermark-based remote digitized court hearing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910079537A CN101827256A (en) | 2009-03-06 | 2009-03-06 | Voiceprint authentication and digital watermark-based remote digitized court hearing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101827256A true CN101827256A (en) | 2010-09-08 |
Family
ID=42690921
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910079537A Pending CN101827256A (en) | 2009-03-06 | 2009-03-06 | Voiceprint authentication and digital watermark-based remote digitized court hearing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101827256A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104160430A (en) * | 2012-02-23 | 2014-11-19 | 锡克拜控股有限公司 | Audible document identification for visually impaired people |
| CN106778186A (en) * | 2017-02-14 | 2017-05-31 | 南方科技大学 | Identity recognition method and device for virtual reality interaction equipment |
| CN107534795A (en) * | 2015-05-15 | 2018-01-02 | 惠普发展公司有限责任合伙企业 | Embed of information into audio stream for connection |
| CN107578777A (en) * | 2016-07-05 | 2018-01-12 | 阿里巴巴集团控股有限公司 | Word-information display method, apparatus and system, audio recognition method and device |
| CN110675117A (en) * | 2019-09-17 | 2020-01-10 | 深圳市天道日新科技有限公司 | Criminal trial network remote litigation method |
| CN115662444A (en) * | 2022-12-14 | 2023-01-31 | 北京惠朗时代科技有限公司 | Electronic seal voice interactive application method and system based on artificial intelligence |
-
2009
- 2009-03-06 CN CN200910079537A patent/CN101827256A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104160430A (en) * | 2012-02-23 | 2014-11-19 | 锡克拜控股有限公司 | Audible document identification for visually impaired people |
| CN107534795A (en) * | 2015-05-15 | 2018-01-02 | 惠普发展公司有限责任合伙企业 | Embed of information into audio stream for connection |
| US10637936B2 (en) | 2015-05-15 | 2020-04-28 | Hewlett-Packard Development Company, L.P. | Embedding information in an audio stream for connectivity |
| CN107578777A (en) * | 2016-07-05 | 2018-01-12 | 阿里巴巴集团控股有限公司 | Word-information display method, apparatus and system, audio recognition method and device |
| CN107578777B (en) * | 2016-07-05 | 2021-08-03 | 阿里巴巴集团控股有限公司 | Text information display method, device and system, and voice recognition method and device |
| CN106778186A (en) * | 2017-02-14 | 2017-05-31 | 南方科技大学 | Identity recognition method and device for virtual reality interaction equipment |
| CN110675117A (en) * | 2019-09-17 | 2020-01-10 | 深圳市天道日新科技有限公司 | Criminal trial network remote litigation method |
| CN115662444A (en) * | 2022-12-14 | 2023-01-31 | 北京惠朗时代科技有限公司 | Electronic seal voice interactive application method and system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108922518B (en) | Voice data amplification method and system | |
| CN101827256A (en) | Voiceprint authentication and digital watermark-based remote digitized court hearing method | |
| CN103310788B (en) | A kind of voice information identification method and system | |
| CN111355781B (en) | Voice information communication management method, device and storage medium | |
| CN101826168A (en) | Remote digital court trial system | |
| Liu et al. | Authentication and recovery algorithm for speech signal based on digital watermarking | |
| CN1808567A (en) | Voice-print authentication device and method of authenticating people presence | |
| CN102708867A (en) | Method and system for identifying faked identity by preventing faked recordings based on voiceprint and voice | |
| DE112012000698B4 (en) | Voice transformation with coded information | |
| Liu et al. | A Spearman correlation coefficient ranking for matching-score fusion on speaker recognition | |
| CN101291239A (en) | Method and apparatus for enhancing effect of meeting | |
| CN101827257A (en) | Iris authentication-based and digital watermark-based remote digital court trial method | |
| WO2016184096A1 (en) | Audio unlocking method and apparatus | |
| CN101826170A (en) | Fingerprint authentication and digital watermark-based remote digital court trial system | |
| CN110827832A (en) | Video identity recognition equipment and method | |
| CN112738338B (en) | Telephone recognition method, device, equipment and medium based on deep learning | |
| CN111275444A (en) | Contract signing-based double recording method and device, terminal and storage medium | |
| CN103778917B (en) | A kind of System and method for that detection identity is pretended to be in phone satisfaction investigation | |
| CN113158148A (en) | Multi-mode digital human copyright management method and system based on block chain | |
| CN113436609A (en) | Voice conversion model and training method thereof, voice conversion method and system | |
| CN109903057B (en) | Contract making method and device | |
| CN117854473B (en) | Zero sample speech synthesis method based on local association information | |
| CN108665901B (en) | Phoneme/syllable extraction method and device | |
| CN102831890A (en) | Method for recognizing text-independent voice prints | |
| CN113571048A (en) | Audio data detection method, device, equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20100908 |