CN101996195B - Searching method and device of voice information in audio files and equipment - Google Patents
Searching method and device of voice information in audio files and equipment Download PDFInfo
- Publication number
- CN101996195B CN101996195B CN2009100916619A CN200910091661A CN101996195B CN 101996195 B CN101996195 B CN 101996195B CN 2009100916619 A CN2009100916619 A CN 2009100916619A CN 200910091661 A CN200910091661 A CN 200910091661A CN 101996195 B CN101996195 B CN 101996195B
- Authority
- CN
- China
- Prior art keywords
- audio file
- audio
- correlation
- degree
- voice messaging
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The invention discloses searching method and device of voice information in audio files and equipment, which are used for realizing the text search on the contents of the audio files, improving the accuracy and the efficiency of the audio file searching and improving the usability of the audio file searching. The searching method comprises the following steps of: carrying out voice identification on each audio file comprising voice information in the audio resource base, converting the audio files into text files comprising text information, and carrying out participle processing on text information of each text file; extracting key words included by corresponding audio files according to words included by each text file, determining the relevance of key words included by each audio file, and establishing an index database of the key words through being combined with the relevance information of each audio file; carrying out specific key word matching in the index database while receiving the voice information searching request carrying the specific key words, and providing the corresponding audio files according to the relevant information of the audio files with the relevance with the specific key words.
Description
Technical field
The present invention relates to the audio search technical field, relate in particular to searching method, device and the equipment of voice messaging in a kind of audio file.
Background technology
Become the information age of geometric growth in quantity of information; Search technique has become one of requisite gordian technique in people's work and the life; Make the information that people can fast search exactly oneself to be needed from the information ocean, thereby greatly improved work and life efficient.Along with search technique reaches its maturity, it is used more and more widely, and people, increase the increasing demand of audio search also in continuous lifting the requirement of search technique.
Existing audio search technology mainly comprises following dual mode:
Mode one, be that audio file adds Word message by manual work in advance, be audio file and set up label, the label of audio file is searched for based on special key words.This mode can't satisfy the demand of audio file being carried out full-text search according to the content of audio file.Simultaneously, because the label of audio file can't be contained the full content of audio file, and label is set up by manual work; Subjective factor is bigger; Cause the accuracy of audio search low, be difficult to guarantee the integrality of Search Results, also can't accurately locate the particular location of special key words in Search Results; If the enormous amount of audio resource storehouse sound intermediate frequency file is huge with making manual work set up the workload of label, cause expending of a large amount of human resources.
Mode two, audio file is searched for based on the audio frequency matching technique; At first need extract the eigenwert of the frequency spectrum or the energy of audio-frequency information to be searched; Extract the eigenwert of the frequency spectrum or the energy of the audio-frequency information of each audio file in the audio resource storehouse then, carry out the coupling of eigenwert at last.The audio frequency matching technique lays particular emphasis on the coupling of the eigenwert of audio frequency itself, and this mode can't satisfy the demand of audio file being carried out full-text search according to the content of audio file equally.Simultaneously; The audio-frequency information that this mode is imported search requires harshness; Not only the content of the content of the audio-frequency information of requirement input and audio resource storehouse sound intermediate frequency file is consistent, but also requires the frequency of audio-frequency information and the frequency and the energy of energy and audio resource storehouse sound intermediate frequency file to be close, the ability successful match; Cause the efficient of audio search low, ease for use is poor.
The audio search technology that provides in the prior art scheme of carrying out full-text search based on the content of audio file is not provided, and the accuracy of audio search is low, efficient is low, ease for use is poor.
Summary of the invention
The present invention provides the searching method and the device of voice messaging in a kind of audio file, in order to realize that the content of audio file is carried out full-text search, improves the accuracy and the efficient of audio search, promotes the ease for use of audio search.
Accordingly, the present invention also provides a kind of terminal device and Website server.
The invention provides the searching method of voice messaging in a kind of audio file, comprising:
To each comprises that the audio file of voice messaging carries out speech recognition in the audio resource storehouse, be converted into the text that comprises Word message, and the Word message of each text is carried out word segmentation processing;
The speech included according to each text extracts the included key word of corresponding audio files; Confirm the degree of correlation of each audio file and included key word; And the relevant information that combines each audio file is set up the index data base of key word, each key word of storage and the degree of correlation of each audio file and the relevant information of each audio file in the said index data base;
When receiving the voice messaging searching request of carrying special key words, in said index data base, carry out the coupling of said special key words, and corresponding audio file is provided according to the relevant information that has an audio file of the degree of correlation with said special key words.
The invention provides the searcher of voice messaging in a kind of audio file, comprising:
Sound identification module is used for that each comprises that the audio file of voice messaging carries out speech recognition to the audio resource storehouse, is converted into the text that comprises Word message, and the Word message of each text is carried out word segmentation processing;
Module set up in index; Be used for extracting the included key word of corresponding audio files according to the included speech of each text; Confirm the degree of correlation of each audio file and included key word, and combine the relevant information of each audio file to set up the index data base of key word;
Index data base is used to store the degree of correlation of each key word and each audio file and the relevant information of each audio file;
The searching disposal module; Be used for when receiving the voice messaging searching request of carrying special key words; In said index data base, carry out the coupling of said special key words, and corresponding audio file is provided according to the relevant information that has an audio file of the degree of correlation with said special key words.
The invention provides a kind of terminal device, comprise the searcher of voice messaging in this audio file.
The invention provides a kind of Website server, comprise the searcher of voice messaging in this audio file.
The searching method of voice messaging, device and equipment in the audio file provided by the invention; To comprise that through speech recognition the audio file of voice messaging is converted into the text that comprises Word message; The text corresponding according to audio file is the full content of audio file, sets up the index data base of key word; When the user imports the search operation of special key words initiation voice messaging; Index data base based on key word provides the audio file that has the degree of correlation with this special key words; Thereby realized the content of audio file is carried out full-text search, remedied the deficiency of existing audio search technology; Because the index data base of key word is set up based on speech recognition technology, and has contained the full content of audio file, thereby has improved the accuracy of audio search, has also improved the efficient of audio search based on the search of key word; When the user initiates to search for, only need the input special key words to get final product, promoted the ease for use of audio search.
Description of drawings
The searcher block diagram of voice messaging in the audio file that Fig. 1 provides for the embodiment of the invention;
The searching method process flow diagram of voice messaging in the audio file that Fig. 2 provides for the embodiment of the invention;
The local search method process flow diagram of voice messaging in the audio file that Fig. 3 provides for embodiment one;
The network search method process flow diagram of voice messaging in the audio file that Fig. 4 provides for embodiment two.
Embodiment
The embodiment of the invention aims to provide a kind of scheme of the content of audio file being carried out full-text search based on key word; Can be according to the special key words of user's input; Content to each audio file in the audio resource storehouse is carried out full-text search, and to the user corresponding audio file is provided.Based on key word the content of audio file is carried out full-text search, can effectively improve the accuracy and the efficient of audio search, promote the ease for use of audio search.
As shown in Figure 1, the embodiment of the invention at first provides the searcher of voice messaging in a kind of audio file, comprising:
Sound identification module 101 is used for that each comprises that the audio file of voice messaging carries out speech recognition to the audio resource storehouse, is converted into the text that comprises Word message, and the Word message of each text is carried out word segmentation processing;
Speech is minimum in the Chinese, independent movable, the significant language element of ability, and speech can comprise a Chinese character, two Chinese characters or a plurality of Chinese characters.Various minutes word algorithms can be realized the word segmentation processing to Word message in the prior art, divide word algorithm mainly to comprise three types: based on the branch word algorithm of string matching, based on the branch word algorithm of understanding with based on the branch word algorithm of adding up;
Module 102 set up in index; Be used for extracting the included key word of corresponding audio files according to the included speech of each text; Confirm the degree of correlation of each audio file and included key word, and combine the relevant information of each audio file to set up the index data base 103 of key word;
Searching disposal module 104; Be used for when receiving the voice messaging searching request of carrying special key words; In index data base 103, carry out the coupling of this special key words, and corresponding audio file is provided according to the relevant information that has an audio file of the degree of correlation with this special key words.
In the practical implementation; In order to promote the degree of accuracy of audio search; When corresponding audio file being provided to the user; The particular location that can also provide this special key words in corresponding audio file, to occur to the user, under this application scenarios, the temporal information that module 102 also combines the included key word of each audio file in this audio file, to occur set up in index when setting up index data base 103; Accordingly, index data base 103 also is used for storing the temporal information that each key word occurs at the audio file with degree of correlation; Searching disposal module 104, also be used for provide have the audio file of the degree of correlation with this special key words in, the temporal information that also provides this special key words in having the audio file of the degree of correlation, to occur.In order accurately to confirm the temporal information that the included key word of each audio file occurs in this audio file; In the practical implementation; Sound identification module 101; The Word message that also is used for each text carries out after the word segmentation processing, and the speech included for each text adds its temporal information that in corresponding audio files, occurs, and is the included speech of each text and adds a timestamp.
In the practical implementation, possibly have the audio file that does not comprise voice messaging in the audio resource storehouse, for example only comprise the audio file of music rhythm, under this application scenarios, the searcher of voice messaging also comprises in this audio file:
Audio frequency parsing module 105 is used for that each audio file of audio resource storehouse is carried out voice and resolves, and extracts the audio file that comprises voice messaging according to the voice analysis result.
Filter out after the audio file that does not comprise voice messaging, can be to each comprises that the audio file of voice messaging carries out speech recognition in the audio resource storehouse.
In the practical implementation, the audio file in the audio resource storehouse may change, and for the accuracy and the completeness that guarantee Search Results, the searcher of voice messaging also comprises in this audio file:
Concrete; If added new audio file in the audio resource storehouse; Then this new audio file is carried out speech recognition, word segmentation processing, keyword extraction; Confirm the degree of correlation of audio file that this is new and included key word, and combine the relevant information of this new audio file in index data base 103, to increase the degree of correlation of this new audio file and included key word and the relevant information of this new audio file; If deleted existing audio file in the audio resource storehouse, then in index data base 103, delete all information relevant with this existing audio file.
The searcher of voice messaging is all applicable to local search and web search in the audio file that the embodiment of the invention provides.If the searcher of voice messaging is arranged in the terminal device that end side is the user in this audio file, can realize that the user carries out local search to the content of each audio file in the local audio resources bank.The local audio resources bank is meant the local storage in user's the terminal device, for example local hard drive, local disk etc.In the local audio resources bank, the relevant information of audio file comprises the file name and the local store path of audio file, and described local store path is " E: music " for example, and expression is stored in local E dish name and is called under the file of " music ".To local search, provide have the audio file of the degree of correlation with this special key words in, the file name and the local store path that have the audio file of the degree of correlation with this special key words also are provided.In the practical implementation; The relevant information of audio file can also comprise other relevant informations such as the size, type, modification time of audio file; Accordingly; Provide have the audio file of the degree of correlation with this special key words in, above-mentioned other relevant information that has the audio file of the degree of correlation with this special key words can also be provided.
Promptly provide in the Website server of the professional website of audio search if the searcher of voice messaging is arranged on network side in this audio file; Through Website server and be installed in cooperatively interacting between the browser of end side, can realize that the user carries out web search to the content of each audio file in the network audio resources bank.The network audio resources bank is meant site databases, and in the network audio resources bank, the relevant information of audio file comprises the file name and the URL (URL) of audio file.To web search, the relevant information that described basis and this special key words have an audio file of the degree of correlation provides corresponding audio file to be meant provides the hyperlink that has the audio file of the degree of correlation with this special key words.
Based on same technical conceive, the embodiment of the invention provides the searching method of voice messaging in a kind of audio file simultaneously, and is as shown in Figure 2, comprising:
S200, each audio file in the audio resource storehouse is carried out voice resolve, extract the audio file that comprises voice messaging according to the voice analysis result;
In the practical implementation,, then need not to carry out this step, directly begin to carry out from S201 if each audio file includes voice messaging in the audio resource storehouse.
S201, to each comprises that the audio file of voice messaging carries out speech recognition in the audio resource storehouse, be converted into the text that comprises Word message, and the Word message of each text carried out word segmentation processing;
In the practical implementation, the Word message of each text being carried out after the word segmentation processing, can also be that the included speech of each text adds its temporal information that in corresponding audio files, occurs.
S202, extract the included key word of corresponding audio files according to the included speech of each text; Confirm the degree of correlation of each audio file and included key word; And the relevant information that combines each audio file is set up the index data base of key word; Accordingly, each key word of storage and the degree of correlation of each audio file and the relevant information of each audio file in the index data base of key word;
In the practical implementation, the degree of correlation of audio file and included key word confirms based on degree of correlation algorithm, and the degree of correlation of audio file and included key word is relevant with the number of times that this key word occurs in audio file, and occurrence number is many more, and the degree of correlation is high more;
In the practical implementation; In order to promote the degree of accuracy of audio search; The temporal information that when setting up the index data base of key word, also combines the included key word of each audio file in this audio file, to occur; Accordingly, also store the temporal information that each key word occurs in the index data base of key word in having the audio file of the degree of correlation.
So far; Accomplished the search preparatory stage of voice messaging in the audio file; In the search preparatory stage, need handle each audio file in the audio resource storehouse, identify voice messaging and convert voice messaging into word information relates based on speech recognition technology; Word message through word segmentation processing and keyword extraction and determine each audio file and the degree of correlation of included key word after set up the index data base of key word.
After the index data base of key word is set up and is accomplished; Can get into the search execute phase of voice messaging in the audio file; The search execute phase is initiated by the user, and through the search operation of input special key words initiation voice messaging, then this method also comprises the steps:
S203, when receiving the voice messaging searching request of carrying special key words; In the index data base of key word, carry out the coupling of this special key words, and corresponding audio file is provided according to the relevant information that has an audio file of the degree of correlation with this special key words;
In the practical implementation, generally from high to low the audio file that has a degree of correlation with this special key words is sorted according to the degree of correlation, the high more ordering of the degree of correlation is forward more;
If also store the temporal information that each key word occurs in the index data base of key word in having the audio file of the degree of correlation; For the ease of the user special key words in the Search Results is accurately located; Provide have the audio file of the degree of correlation with this special key words in; The temporal information that also provides special key words in having the audio file of the degree of correlation, to occur, specifically the form with time shaft provides.
In the practical implementation, also comprise regularly or the audio file in the audio resource storehouse when changing, the index data base of key word is carried out updating steps.
To be example with local search and web search respectively below, the search plan of voice messaging in the audio file that the detailed description embodiment of the invention provides.
Embodiment one
Present embodiment provides the local search scheme of voice messaging in the audio file; Corresponding audio resources bank (can be called the local audio resources bank) is arranged on end side; Be specially the local storage in user's the terminal device; In order to realize local search, the searcher of voice messaging in the audio file that the embodiment of the invention provides need be set in user's terminal device to voice messaging in the audio file.The local search flow process of voice messaging is as shown in Figure 3 in the audio file, comprises local search preparatory stage and local search execute phase.The local search preparatory stage, comprise the steps:
S301, terminal device extract a untreated audio file from the audio resource storehouse, current audio file is carried out voice resolve;
S302, terminal device judge according to the voice analysis result whether current audio file comprises voice messaging, if if then carry out S303 not, then turn to and carry out S307;
S303, terminal device carry out speech recognition to current audio file, are converted into the text that comprises Word message;
S304, terminal device carry out word segmentation processing to the Word message of current text, and are included its temporal information that in corresponding audio files, occurs of speech interpolation of current text;
S305, terminal device extract the included key word of corresponding audio files according to the included speech of current text, confirm the degree of correlation of current audio file and included key word;
S306, terminal device store the temporal information that the file name of the degree of correlation of current audio file and included key word, current audio file and local store path and the current included key word of audio file occur in the index data base of key word in this audio file;
S307, the current audio file of terminal device are set to handle;
S308, terminal device judge whether also there is untreated audio file in the audio resource storehouse, if then return and carry out S301; If not; Then the index data base of key word set up to be accomplished, and promptly the local search preparatory stage accomplishes, and follow-uply can get into the local search execute phase.
If the user imports special key words in the local search toolbar, initiate the local search of voice messaging, then the local search execute phase, comprise the steps:
S309, when receiving the local search query of the voice messaging that carries special key words, terminal device carries out the coupling of this special key words in the index data base of key word;
S310, terminal device basis and this special key words have the file name and the local store path of the audio file of the degree of correlation; The temporal information that provides corresponding audio file and this special key words in having the audio file of the degree of correlation, to occur can also provide the file name and the local store path of this audio file certainly in the lump;
Accordingly, the temporal information that audio file and this special key words occur in having the audio file of the degree of correlation, the file name of this audio file and local store path represent the confession user and check on terminal device.
It is to be noted; In the practical implementation since the local audio resources bank in audio file can change; For example the user has added new audio file or has deleted existing audio file in the local storage in the local storage of terminal device; Therefore need regularly or the audio file in the local audio resources bank when changing, the index data base of key word is upgraded, to guarantee the accuracy and the completeness of local search results.
Embodiment two
Present embodiment provides the web search scheme of voice messaging in the audio file.Corresponding audio resources bank (can be called the local audio resources bank) is arranged on network side; Be specially site databases; In order to realize web search, the searcher of voice messaging in the audio file that the embodiment of the invention provides need be set in the Website server that the professional website of audio search is provided to voice messaging in the audio file.The web search flow process of voice messaging is as shown in Figure 4 in the audio file, comprises web search preparatory stage and web search execute phase.The web search preparatory stage, comprise the steps:
S401, Website server extract a untreated audio file from the audio resource storehouse, current audio file is carried out voice resolve;
S402, Website server judge according to the voice analysis result whether current audio file comprises voice messaging, if, then carry out S403, if not, then turn to and carry out S407;
S403, Website server carry out speech recognition to current audio file, are converted into the text that comprises Word message;
S404, Website server carry out word segmentation processing to the Word message of current text, and are included its temporal information that in corresponding audio files, occurs of speech interpolation of current text;
S405, Website server extract the included key word of corresponding audio files according to the included speech of current text, confirm the degree of correlation of current audio file and included key word;
S406, Website server store the temporal information that the file name of the degree of correlation of current audio file and included key word, current audio file and URL and the current included key word of audio file occur in the index data base of key word in this audio file
S407, the current audio file of Website server are set to handle;
S408, Website server judge whether also there is untreated audio file in the audio resource storehouse, if then return and carry out S401; If not; Then the index data base of key word set up to be accomplished, and promptly the web search preparatory stage accomplishes, and follow-uply can get into the web search execute phase.
If the user imports special key words in the cyber stalker hurdle of the browser of end side, initiate the web search of voice messaging, then the web search execute phase, comprise the steps:
S409, when receiving the network search request of the voice messaging that carries special key words, Website server carries out the coupling of this special key words in the index data base of key word;
S410, Website server basis and this special key words have the file name and the URL of the audio file of the degree of correlation, and the hyperlink of corresponding audio file and the temporal information that this special key words occurs in having the audio file of the degree of correlation are provided;
Accordingly, the temporal information that the hyperlink of audio file and this special key words occur in having the audio file of the degree of correlation sends to the browser of end side through transmission network, on terminal device, represents to supply the user to check.
It is to be noted; In the practical implementation since the network audio resources bank in audio file can change; For example add new audio file in the site databases or deleted existing audio file; Therefore need regularly or the audio file in the network audio resources bank when changing, the index data base of key word is upgraded, to guarantee the accuracy and the completeness of web search results.
The searching method of voice messaging, device and equipment in the audio file provided by the invention; To comprise that through speech recognition the audio file of voice messaging is converted into the text that comprises Word message; The text corresponding according to audio file is the full content of audio file, sets up the index data base of key word; When the user imports the search of special key words initiation voice messaging; Index data base based on key word provides the audio file that has the degree of correlation with this special key words; Thereby realized the content of audio file is carried out full-text search, remedied the deficiency of existing audio search technology; Because the index data base of key word is set up based on speech recognition technology, and has contained the full content of audio file, thereby has improved the accuracy of audio search, has also improved the efficient of audio search based on the search of key word; When the user initiates to search for, only need the input special key words to get final product, promoted the ease for use of audio search.
The searching method of voice messaging, device and equipment in the audio file provided by the invention; In the index data base of key word, also store each key word and have the temporal information that occurs in the audio file of the degree of correlation; When the user imports the search of special key words initiation voice messaging; The temporal information that can also provide this special key words in having the audio file of the degree of correlation, to occur based on the index data base of key word, thus realized the particular location of accurate location special key words in Search Results.
It will be understood by those skilled in the art that embodiments of the invention can be provided as method, device, equipment or computer program.Therefore, the present invention can adopt the form of the embodiment of complete hardware embodiment, complete software implementation example or combination software and hardware aspect.And the present invention can be employed in the form that one or more computer-usable storage medium (including but not limited to magnetic disk memory, CD-ROM, optical memory etc.) that wherein include computer usable program code go up the computer program of implementing.
The present invention is that reference is described according to the process flow diagram and/or the block scheme of method, device, equipment and the computer program of the embodiment of the invention.Should understand can be by the flow process in each flow process in computer program instructions realization flow figure and/or the block scheme and/or square frame and process flow diagram and/or the block scheme and/or the combination of square frame.Can provide these computer program instructions to the processor of multi-purpose computer, special purpose computer, Embedded Processor or other programmable data processing device to produce a machine, make the instruction of carrying out through the processor of computing machine or other programmable data processing device produce to be used for the device of the function that is implemented in flow process of process flow diagram or a plurality of flow process and/or square frame of block scheme or a plurality of square frame appointments.
These computer program instructions also can be stored in ability vectoring computer or the computer-readable memory of other programmable data processing device with ad hoc fashion work; Make the instruction that is stored in this computer-readable memory produce the manufacture that comprises command device, this command device is implemented in the function of appointment in flow process of process flow diagram or a plurality of flow process and/or square frame of block scheme or a plurality of square frame.
These computer program instructions also can be loaded on computing machine or other programmable data processing device; Make on computing machine or other programmable devices and to carry out the sequence of operations step producing computer implemented processing, thereby the instruction of on computing machine or other programmable devices, carrying out is provided for being implemented in the step of the function of appointment in flow process of process flow diagram or a plurality of flow process and/or square frame of block scheme or a plurality of square frame.
Although described the preferred embodiments of the present invention, in a single day those skilled in the art get the basic inventive concept could of cicada, then can make other change and modification to these embodiment.So accompanying claims is intended to be interpreted as all changes and the modification that comprises preferred embodiment and fall into the scope of the invention.
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, belong within the scope of claim of the present invention and equivalent technologies thereof if of the present invention these are revised with modification, then the present invention also is intended to comprise these changes and modification interior.
Claims (15)
1. the searching method of voice messaging in the audio file is characterized in that, comprising:
To each comprises that the audio file of voice messaging carries out speech recognition in the audio resource storehouse, be converted into the text that comprises Word message, and the Word message of each text is carried out word segmentation processing;
The speech included according to each text extracts the included key word of corresponding audio files; Confirm the degree of correlation of each audio file and included key word; And the relevant information that combines each audio file is set up the index data base of key word, each key word of storage and the degree of correlation of each audio file and the relevant information of each audio file in the said index data base;
When receiving the voice messaging searching request of carrying special key words, in said index data base, carry out the coupling of said special key words, and corresponding audio file is provided according to the relevant information that has an audio file of the degree of correlation with said special key words.
2. the method for claim 1; It is characterized in that; The temporal information that when setting up said index data base, also combines the included key word of each audio file in this audio file, to occur is also stored the temporal information that each key word occurs in having the audio file of the degree of correlation in the said index data base; And
Provide have the audio file of the degree of correlation with said special key words in, the temporal information that also provides said special key words in having the audio file of the degree of correlation, to occur.
3. method as claimed in claim 2 is characterized in that, also comprises:
The Word message of each text is carried out after the word segmentation processing, is that the included speech of each text adds its temporal information that in corresponding audio files, occurs.
4. like claim 1,2 or 3 arbitrary described methods, it is characterized in that, from high to low the audio file that has a degree of correlation with said special key words is sorted according to the degree of correlation.
5. the method for claim 1; It is characterized in that; Said audio resource lab setting is in end side, and said voice messaging searching request is the local search query of voice messaging, and the relevant information of said audio file comprises the file name and the local store path of audio file; And
Provide have the audio file of the degree of correlation with said special key words in, the file name and the local store path that have the audio file of the degree of correlation with said special key words also are provided.
6. the method for claim 1; It is characterized in that; Said audio resource lab setting is at network side, and said voice messaging searching request is the network search request of voice messaging, and the relevant information of said audio file comprises the file name and the uniform resource position mark URL of audio file; And
The relevant information that said basis and said special key words have an audio file of the degree of correlation provides corresponding audio file to be meant provides the hyperlink that has the audio file of the degree of correlation with said special key words.
7. the method for claim 1 is characterized in that, each comprises that the audio file of voice messaging carries out also comprising before the speech recognition in to the audio resource storehouse:
Each audio file in the audio resource storehouse is carried out voice resolve, extract the audio file that comprises voice messaging according to the voice analysis result.
8. the method for claim 1 is characterized in that, also comprises:
When regularly perhaps the audio file in said audio resource storehouse changes, said index data base is upgraded.
9. the searcher of voice messaging in the audio file is characterized in that, comprising:
Sound identification module is used for that each comprises that the audio file of voice messaging carries out speech recognition to the audio resource storehouse, is converted into the text that comprises Word message, and the Word message of each text is carried out word segmentation processing;
Module set up in index; Be used for extracting the included key word of corresponding audio files according to the included speech of each text; Confirm the degree of correlation of each audio file and included key word, and combine the relevant information of each audio file to set up the index data base of key word;
Index data base is used to store the degree of correlation of each key word and each audio file and the relevant information of each audio file;
The searching disposal module; Be used for when receiving the voice messaging searching request of carrying special key words; In said index data base, carry out the coupling of said special key words, and corresponding audio file is provided according to the relevant information that has an audio file of the degree of correlation with said special key words.
10. device as claimed in claim 9 is characterized in that,
The temporal information that module also combines the included key word of each audio file in this audio file, to occur set up in said index when setting up said index data base;
Said index data base also is used for storing the temporal information that each key word occurs at the audio file with degree of correlation;
Said searching disposal module, also be used for provide have the audio file of the degree of correlation with said special key words in, the temporal information that also provides said special key words in having the audio file of the degree of correlation, to occur.
11. device as claimed in claim 10 is characterized in that,
Said sound identification module, the Word message that also is used for each text carries out after the word segmentation processing, is that the included speech of each text adds its temporal information that in corresponding audio files, occurs.
12. device as claimed in claim 9 is characterized in that, also comprises:
The audio frequency parsing module; Be used at sound identification module before each comprises that the audio file of voice messaging carries out speech recognition to the audio resource storehouse; Each audio file in the audio resource storehouse is carried out voice resolve, extract the audio file that comprises voice messaging according to the voice analysis result.
13. device as claimed in claim 9 is characterized in that, also comprises:
Update module is used for regularly or the audio file in said audio resource storehouse when changing, and said index data base is upgraded.
14. a terminal device is characterized in that, comprises like the arbitrary described searcher of claim 9 to 13.
15. a Website server is characterized in that, comprises like the arbitrary described searcher of claim 9 to 13.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100916619A CN101996195B (en) | 2009-08-28 | 2009-08-28 | Searching method and device of voice information in audio files and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100916619A CN101996195B (en) | 2009-08-28 | 2009-08-28 | Searching method and device of voice information in audio files and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101996195A CN101996195A (en) | 2011-03-30 |
| CN101996195B true CN101996195B (en) | 2012-07-11 |
Family
ID=43786362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100916619A Expired - Fee Related CN101996195B (en) | 2009-08-28 | 2009-08-28 | Searching method and device of voice information in audio files and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101996195B (en) |
Families Citing this family (43)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102867511A (en) * | 2011-07-04 | 2013-01-09 | 余喆 | Method and device for recognizing natural speech |
| CN102867512A (en) * | 2011-07-04 | 2013-01-09 | 余喆 | Method and device for recognizing natural speech |
| CN102316361B (en) * | 2011-07-04 | 2014-05-21 | 深圳市车音网科技有限公司 | Audio-frequency / video-frequency on demand method based on natural speech recognition and system thereof |
| CN103425668A (en) * | 2012-05-16 | 2013-12-04 | 联想(北京)有限公司 | Information search method and electronic equipment |
| CN103218454A (en) * | 2013-05-06 | 2013-07-24 | 百度在线网络技术（北京）有限公司 | Voice-data-based file searching method, voice-data-based file device and voice-data-based file system |
| CN103365959A (en) * | 2013-06-03 | 2013-10-23 | 深圳市爱渡飞科技有限公司 | Voice search method and voice search device |
| CN103366010A (en) * | 2013-07-25 | 2013-10-23 | 北京小米科技有限责任公司 | Method and device for searching audio file |
| CN104375997A (en) * | 2013-08-13 | 2015-02-25 | 腾讯科技（深圳）有限公司 | Method and device for adding note information to instant messaging audio information |
| CN104424343A (en) * | 2013-09-11 | 2015-03-18 | 中兴通讯股份有限公司 | Information query method and terminal device |
| CN104572714A (en) * | 2013-10-18 | 2015-04-29 | 英业达科技有限公司 | Learning video inquiring system and learning video inquiring method |
| CN104809115A (en) * | 2014-01-24 | 2015-07-29 | 贝壳网际（北京）安全技术有限公司 | Searching method and terminal device |
| CN104978366B (en) * | 2014-04-14 | 2018-09-25 | 深圳市北科瑞声科技股份有限公司 | Voice data index establishing method based on mobile terminal and system |
| CN104391924A (en) * | 2014-11-21 | 2015-03-04 | 南京讯思雅信息科技有限公司 | Mixed audio and video search method and system |
| CN105760399A (en) * | 2014-12-19 | 2016-07-13 | 华为软件技术有限公司 | Data retrieval method and device |
| CN104834740A (en) * | 2015-05-20 | 2015-08-12 | 深圳市东方泰明科技有限公司 | Full-automatic audio/video structuralized accurate searching method |
| CN105550217B (en) * | 2015-12-03 | 2021-05-07 | 腾讯科技（深圳）有限公司 | Scene music searching method and scene music searching device |
| CN105551488A (en) * | 2015-12-15 | 2016-05-04 | 深圳Tcl数字技术有限公司 | Voice control method and system |
| CN105824930A (en) * | 2016-03-17 | 2016-08-03 | 深圳市金立通信设备有限公司 | Voice message processing method and terminal |
| CN106024013B (en) * | 2016-04-29 | 2022-01-14 | 努比亚技术有限公司 | Voice data searching method and system |
| CN106202204A (en) * | 2016-06-24 | 2016-12-07 | 维沃移动通信有限公司 | The lookup method of a kind of voice document and mobile terminal |
| CN108121715B (en) * | 2016-11-28 | 2022-01-25 | 中国移动通信集团公司 | Character labeling method and character labeling device |
| CN108170691A (en) * | 2016-12-07 | 2018-06-15 | 北京国双科技有限公司 | It is associated with the determining method and apparatus of document |
| CN107016109B (en) * | 2017-04-14 | 2018-11-30 | 维沃移动通信有限公司 | A kind of photo film making method and mobile terminal |
| CN106911832B (en) * | 2017-04-28 | 2020-06-02 | 四川音创伟业科技有限公司 | Voice recording method and device |
| CN109559764A (en) * | 2017-09-27 | 2019-04-02 | 北京国双科技有限公司 | The treating method and apparatus of audio file |
| CN108257597A (en) * | 2017-12-28 | 2018-07-06 | 合肥凯捷技术有限公司 | A kind of audio retrieval system based on speech recognition |
| CN108320318B (en) * | 2018-01-15 | 2023-07-28 | 腾讯科技（深圳）有限公司 | Image processing method, device, computer equipment and storage medium |
| CN109741750A (en) * | 2018-05-09 | 2019-05-10 | 北京字节跳动网络技术有限公司 | A kind of method of speech recognition, document handling method and terminal device |
| CN108829765A (en) * | 2018-05-29 | 2018-11-16 | 平安科技（深圳）有限公司 | A kind of information query method, device, computer equipment and storage medium |
| CN108833971A (en) * | 2018-06-06 | 2018-11-16 | 北京奇艺世纪科技有限公司 | A kind of method for processing video frequency and device |
| CN109034418B (en) * | 2018-07-26 | 2021-05-28 | 国家电网公司 | Operation site information transmission method and system |
| CN109299227B (en) * | 2018-11-07 | 2023-06-02 | 平安医疗健康管理股份有限公司 | Information query method and device based on voice recognition |
| CN109274586A (en) * | 2018-11-14 | 2019-01-25 | 深圳市云歌人工智能技术有限公司 | Storage method, device and the storage medium of chat message |
| CN109460209B (en) * | 2018-12-20 | 2022-03-01 | 广东小天才科技有限公司 | Control method for dictation and reading progress and electronic equipment |
| CN111353065A (en) * | 2018-12-20 | 2020-06-30 | 北京嘀嘀无限科技发展有限公司 | Voice archive storage method, device, equipment and computer readable storage medium |
| CN110335598A (en) * | 2019-06-26 | 2019-10-15 | 重庆金美通信有限责任公司 | A kind of wireless narrow band channel speech communication method based on speech recognition |
| CN110287364B (en) * | 2019-06-28 | 2021-10-08 | 合肥讯飞读写科技有限公司 | Voice search method, system, device and computer readable storage medium |
| CN110275978A (en) * | 2019-07-01 | 2019-09-24 | 成都启英泰伦科技有限公司 | Quick storage of the voice big data on redundant arrays of inexpensive disks and access amending method |
| CN110275979A (en) * | 2019-07-01 | 2019-09-24 | 成都启英泰伦科技有限公司 | A kind of mapping management process of voice data and text data |
| CN111008300A (en) * | 2019-11-20 | 2020-04-14 | 四川互慧软件有限公司 | Keyword-based timestamp positioning search method in audio and video |
| CN111161738A (en) * | 2019-12-27 | 2020-05-15 | 苏州欧孚网络科技股份有限公司 | Voice file retrieval system and retrieval method thereof |
| CN112115282A (en) * | 2020-09-17 | 2020-12-22 | 北京达佳互联信息技术有限公司 | Question answering method, device, equipment and storage medium based on search |
| CN113299279A (en) * | 2021-05-18 | 2021-08-24 | 上海明略人工智能(集团)有限公司 | Method, apparatus, electronic device and readable storage medium for associating voice data and retrieving voice data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1296257A (en) * | 1999-11-10 | 2001-05-23 | 国际商业机器公司 | Automatic index based on semantic unit in data file system and searching method and equipment |
| CN101281534A (en) * | 2008-05-28 | 2008-10-08 | 叶睿智 | Method for searching multimedia resource based on audio content retrieval |
| CN101305360A (en) * | 2005-11-08 | 2008-11-12 | 微软公司 | Indexing and searching speech with text meta-data |
| CN101351838A (en) * | 2005-12-30 | 2009-01-21 | 坦德伯格电信公司 | Searchable multimedia stream |
| CN101510222A (en) * | 2009-02-20 | 2009-08-19 | 北京大学 | Multilayer index voice document searching method and system thereof |
-
2009
- 2009-08-28 CN CN2009100916619A patent/CN101996195B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1296257A (en) * | 1999-11-10 | 2001-05-23 | 国际商业机器公司 | Automatic index based on semantic unit in data file system and searching method and equipment |
| CN101305360A (en) * | 2005-11-08 | 2008-11-12 | 微软公司 | Indexing and searching speech with text meta-data |
| CN101351838A (en) * | 2005-12-30 | 2009-01-21 | 坦德伯格电信公司 | Searchable multimedia stream |
| CN101281534A (en) * | 2008-05-28 | 2008-10-08 | 叶睿智 | Method for searching multimedia resource based on audio content retrieval |
| CN101510222A (en) * | 2009-02-20 | 2009-08-19 | 北京大学 | Multilayer index voice document searching method and system thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101996195A (en) | 2011-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101996195B (en) | Searching method and device of voice information in audio files and equipment | |
| CN100458795C (en) | Intelligent word input method and input method system and updating method thereof | |
| CN102023989B (en) | Information retrieval method and system thereof | |
| CN103425687A (en) | Retrieval method and system based on queries | |
| CN103092943B (en) | A kind of method of advertisement scheduling and advertisement scheduling server | |
| CN105224554A (en) | Search word is recommended to carry out method, system, server and the intelligent terminal searched for | |
| CN102591880A (en) | Information providing method and device | |
| CN101499098A (en) | Web page assessed value confirming and employing method and system | |
| CN102722499B (en) | Search engine and implementation method thereof | |
| CN102063469A (en) | Method and device for acquiring relevant keyword message and computer equipment | |
| CN101179472A (en) | Network resource searching method and searching system | |
| CN102193917A (en) | Method and device for processing and querying data | |
| CN103810224A (en) | Information persistence and query method and device | |
| CN104850554A (en) | Searching method and system | |
| CN102722501A (en) | Search engine and realization method thereof | |
| CN106844640A (en) | A kind of web data analysis and processing method | |
| CN102737021A (en) | Search engine and realization method thereof | |
| JP2018501540A (en) | Stopword identification method and apparatus | |
| CN105302807A (en) | Method and apparatus for obtaining information category | |
| CN105354318A (en) | File searching method and device | |
| CN102436458B (en) | A kind of method of command analysis and system thereof | |
| CN104484413A (en) | Method and device for obtaining searching results | |
| US8655886B1 (en) | Selective indexing of content portions | |
| CN105512270A (en) | Method and device for determining related objects | |
| CN105653533A (en) | Method and device for updating classified associated word set |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120711 Termination date: 20210828 |