CN113312541A - Voice search method, device and storage medium - Google Patents
Voice search method, device and storage medium Download PDFInfo
- Publication number
- CN113312541A CN113312541A CN202110871873.XA CN202110871873A CN113312541A CN 113312541 A CN113312541 A CN 113312541A CN 202110871873 A CN202110871873 A CN 202110871873A CN 113312541 A CN113312541 A CN 113312541A
- Authority
- CN
- China
- Prior art keywords
- target
- keyword set
- keyword
- user
- voice
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9532—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application relates to the technical field of internet, and the embodiment of the application discloses a voice search method, a device and a storage medium, wherein the method comprises the following steps: acquiring target voice input by a user; converting the target voice into a target text; extracting features of the target text to obtain a first keyword set; acquiring a target browsing record of the user in a preset time period; extracting features of the target browsing record to obtain a second keyword set; determining a third keyword set according to the first keyword set and the second keyword set; and searching according to the third keyword set to obtain a target search result set. By adopting the embodiment of the application, the voice search precision can be improved.
Description
Technical Field
The present application relates to the field of internet technologies, and in particular, to a voice search method, apparatus, and storage medium.
Background
With the rapid development of electronic technology, the functions of electronic devices (such as mobile phones) are also becoming more and more powerful, voice search also becomes a necessary tool for users, and users can search data required by various users through the voice search function, for example, various data such as food information and character information.
At present, searching is usually performed based on a recognition result of a voice content of a user, the searching precision often depends on the recognition precision of the voice content, sometimes, a pronunciation of the user is inaccurate, or a voice recognition is wrong, so that a searching result is also inaccurate, and therefore, a problem of how to improve the voice searching precision needs to be solved urgently.
Disclosure of Invention
The embodiment of the application provides a voice search method, a voice search device and a storage medium, which can improve the voice search precision.
In a first aspect, an embodiment of the present application provides a voice search method, where the method includes:
acquiring target voice input by a user;
converting the target voice into a target text;
extracting features of the target text to obtain a first keyword set;
acquiring a target browsing record of the user in a preset time period;
extracting features of the target browsing record to obtain a second keyword set;
determining a third keyword set according to the first keyword set and the second keyword set;
and searching according to the third keyword set to obtain a target search result set.
In a second aspect, an embodiment of the present application provides a voice search apparatus, where the apparatus includes: a first obtaining unit, a converting unit, a first extracting unit, a second obtaining unit, a second extracting unit, a determining unit and a searching unit, wherein,
the first acquisition unit is used for acquiring target voice input by a user;
the conversion unit is used for converting the target voice into a target text;
the first extraction unit is used for extracting the characteristics of the target text to obtain a first keyword set;
the second obtaining unit is used for obtaining a target browsing record of the user in a preset time period;
the second extraction unit is used for extracting features of the target browsing record to obtain a second keyword set;
the determining unit is used for determining a third keyword set according to the first keyword set and the second keyword set;
and the searching unit is used for searching according to the third key word set to obtain a target searching result set.
In a third aspect, an embodiment of the present application provides a server, including a processor, a memory, a communication interface, and one or more programs, where the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the first aspect of the embodiment of the present application.
In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program for electronic data exchange, where the computer program enables a computer to perform some or all of the steps described in the first aspect of the embodiment of the present application.
In a fifth aspect, embodiments of the present application provide a computer program product, where the computer program product includes a non-transitory computer-readable storage medium storing a computer program, where the computer program is operable to cause a computer to perform some or all of the steps as described in the first aspect of the embodiments of the present application. The computer program product may be a software installation package.
The embodiment of the application has the following beneficial effects:
it can be seen that the voice search method, apparatus, and storage medium described in the embodiments of the present application obtain a target voice input by a user, convert the target voice into a target text, perform feature extraction on the target text to obtain a first keyword set, obtain a target browsing record of the user in a preset time period, perform feature extraction on the target browsing record to obtain a second keyword set, determine a third keyword set according to the first keyword set and the second keyword set, perform search according to the third keyword set to obtain a target search result set, so that the voice can be converted into a text and corresponding keywords can be extracted, and keywords in the browsing record can be extracted, and then the keywords of the text and the keywords can be integrated, where the integrated keywords include a user search appeal expressed by the voice of the user and also include keywords that may be inspired by the user in the browsing record, the third key word has stronger user intention, and is more favorable for obtaining a search result required by the user in the search process.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flowchart of a voice search method provided in an embodiment of the present application;
FIG. 2 is a schematic illustration of an interface demonstration of a voice search provided by an embodiment of the present application;
FIG. 3 is a schematic view illustrating a scene of a voice search method according to an embodiment of the present application;
FIG. 4 is a flow chart illustrating another speech searching method provided by an embodiment of the present application;
fig. 5 is a schematic structural diagram of a server provided in an embodiment of the present application;
fig. 6 is a block diagram illustrating functional units of a speech search apparatus according to an embodiment of the present application.
Detailed Description
In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms "first," "second," and the like in the description and claims of the present application and in the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
The following describes embodiments of the present application in detail.
Referring to fig. 1, fig. 1 is a schematic flow chart of a voice search method according to an embodiment of the present application, where the voice search method includes:
101. and acquiring target voice input by a user.
In specific implementation, as shown in fig. 2, a voice control can be displayed on a search interface of the electronic device, a user can input a voice through the voice control, and then can input a target voice, and then the server can obtain the target voice input by the user, or the user can load an audio file from a local or cloud storage, and the voice contained in the audio file is used as the target voice. As shown in fig. 3, the user device and the server may maintain communication, and further, the user may perform voice input through the user device (e.g., a mobile phone) to obtain a target voice, and then transmit the target voice to the server, and the server may complete a search process based on the target voice.
102. And converting the target voice into a target text.
The target voice can be preprocessed, and then the preprocessed target voice is converted into a target text, wherein the preprocessing includes at least one of the following: signal noise reduction, signal filtering, signal amplification, etc., without limitation. In this embodiment of the application, the target text may be a text in a preset language, and the preset language may include at least one of the following: chinese, english, russian, french, ethnic minority languages (e.g., tibetan), local languages, etc., without limitation.
Optionally, in the step 102, converting the target speech into the target text may include the following steps:
21. identifying the target voice to obtain a target language type;
22. determining a target voice recognition model corresponding to the target voice type according to a mapping relation between a preset language type and a voice recognition model;
23. and converting the target voice into a target text through the target voice recognition model.
In specific implementation, a mapping relationship between a preset language type and a speech recognition model may be pre-stored, that is, different languages may select different speech recognition models, a function of the speech recognition model is used to convert speech into text, and the speech recognition model may be any model having the function.
Specifically, the target speech can be recognized to obtain the target language type, the recognition can be performed specifically through the characteristics of the speech, such as tone, pronunciation and the like, and then the target speech recognition model corresponding to the target speech type is determined according to the mapping relation between the preset language type and the speech recognition model, and then the target speech can be converted into the target text through the target speech recognition model, so that the corresponding speech recognition model can be selected based on the speech characteristics of the user, and the speech recognition precision is favorably improved.
103. And extracting the characteristics of the target text to obtain a first keyword set.
The first keyword set may include at least one keyword, and the keyword includes content that the user needs to search. The feature extraction may be a keyword extraction technique. In specific implementation, keyword extraction can be performed on the target text to obtain a first keyword set.
104. And acquiring a target browsing record of the user in a preset time period.
The preset time period can be set by the user or defaulted by the system, the target browsing record can be a webpage browsed by the user, or the content watched by the eyes of the user can be acquired by an eyeball tracking technology. The browsing record indicates the thought track of the user, under the condition that the user sees what under the browsing condition, or suddenly thinks what, the user can initiate the search behavior, and further, the server can acquire the browsing record of the user in a preset time period under the authorization condition of the user, and in the specific implementation, the browsing record closer to the current time has more extracted keywords, or the extracted keywords have higher importance.
105. And extracting features of the target browsing record to obtain a second keyword set.
The second keyword set may include at least one keyword, and the feature extraction may also be a keyword extraction technique. In specific implementation, feature extraction may be performed on the target browsing record, and specifically, keyword extraction may be performed on the target browsing record to obtain the second keyword set. Since the browsing history covers the user's will, the keywords in the second keyword set also carry the user's search will.
106. And determining a third keyword set according to the first keyword set and the second keyword set.
The first keyword set and the second keyword set may have the same, homophonic, synonymous and near-synonym conditions, under these conditions, corresponding keywords of the two keyword sets can be considered to be related to each other, for example, a certain keyword in the second keyword set is the same as a certain keyword in the first keyword set, and then the two keywords are related to each other.
Optionally, in step 106, determining a third keyword set according to the first keyword set and the second keyword set may include the following steps:
61. selecting keywords related to the keywords in the first keyword set from the second keyword set to obtain at least one keyword;
62. merging the second keyword set and the at least one keyword to obtain a fourth keyword set;
63. and selecting a preset number of keywords from the fourth keyword set as the third keyword set.
The preset number may be set by the user or default to the system, for example, the preset number may be 2 or 3.
Specifically, keywords related to the keywords in the first keyword set may be selected from the second keyword set to obtain at least one keyword, and the correlation may be understood as a correlation, for example, a correlation between homophones, a correlation between the same keywords, and a correlation between synonymous or near-synonymous keywords.
Furthermore, the second keyword set and at least one keyword may be merged to obtain a fourth keyword set, for example, only one keyword may be retained for the same keyword, and for example, if the keyword is similar, the keyword may be selected according to the heat degree of the word, and for example, if the keyword is of a same tone and different meaning, there may be a voice recognition error, the corresponding keyword in the browsing record may be used as the standard, a predetermined number of keywords may be selected from the fourth keyword set as the third keyword set, if the number of keywords is too many, the search result may be too few, the search result really required by the user is omitted, a certain number of keywords may be retained, the predetermined number of keywords may be randomly selected, or the previous predetermined number of keywords may be selected according to the heat degree of the keywords, so that the number of the obtained search results is moderate, and the probability of the search result required by the user is reduced, and search results can be enriched to a certain extent.
In addition, in a possible example, during the step 62, the first keyword set may also be merged with at least one keyword to obtain a fourth keyword set, and a specific merging manner is similar to that described above, and is not described herein again.
Optionally, before step 106, the following steps may be further included:
a1, acquiring target text search information;
a2, performing feature extraction on the target text search information to obtain a fifth keyword set;
and determining a third keyword set according to the first keyword set, the second keyword set and the fifth keyword set.
In the specific implementation, target text search information can be obtained, the target text search information can be manually input by a user, feature extraction can be performed on the target text search information to obtain a fifth keyword set, the fifth keyword set comprises at least one keyword, the feature extraction technical means can be a keyword extraction technology, sometimes, some word users do not know how to pronounce the words, the words can be input through the text, and then the third keyword set is determined according to the first keyword set, the second keyword set and the fifth keyword set, namely, keywords from different sources all express user requirements, and then the keywords can be integrated, so that the keywords are more suitable for the user requirements, and the voice search precision is improved.
107. And searching according to the third keyword set to obtain a target search result set.
In a specific implementation, the target search result set may include at least one search result, and since the third keyword set includes a user search appeal expressed by the user's voice and also includes keywords which may be inspired by the user in the browsing record, such a third keyword has a strong user intention, and is more beneficial to obtaining a search result required by the user in the search process.
Optionally, in step 107, searching according to the third keyword set to obtain a target search result set, the method may include the following steps:
71. searching according to the third keyword set to obtain an initial search result set;
72. selecting P search results which are ranked at the top from the initial search result set, wherein P is an integer larger than 1;
73. acquiring the age of the user to obtain a target age;
74. and screening and sequencing the P search results according to the target age to obtain the target search result set.
In a specific implementation, the initial search result set includes at least one search result, the server may search in the database according to the third keyword set to obtain an initial search result set, the initial search result set is only one ranking display result obtained by searching according to a search algorithm, P search results ranked at the top in the initial search result set may be selected, P is an integer greater than 1, and the higher the degree of association with the keywords is, that is, the higher the search accuracy is.
Further, the age of the user may be obtained, for example, when the user is a login user, the age of the user may be obtained according to the registration information of the user, and for example, the age of the user may be identified by the target voice to obtain the target age, and the voice pronunciations of the users are also different due to different age stages. And the preferences of the users on the search results are different at different ages, so that the P search results can be screened and sorted by age to obtain a target search result set, and the target search result set better meets the age requirements of the users and is beneficial to improving the satisfaction of the users.
Further, optionally, in the step 74, the step of screening and sorting the P search results according to the target age to obtain the target search result set may include the following steps:
741. acquiring browsing records of each search result in the P search results to obtain P browsing records, wherein each browsing record corresponds to the age information of the user;
742. determining a browsing distribution histogram corresponding to each search result in the P search results according to the P browsing records to obtain P browsing distribution histograms, wherein the horizontal axis of the browsing distribution histogram is age, and the vertical axis of the browsing distribution histogram is user number;
743. determining the association degree between each search result in the P search results and the target age according to the P browsing distribution histograms to obtain P association degrees;
743. screening out the relevance in a preset range from the P relevance to obtain Q relevance, wherein Q is a positive integer less than or equal to P;
744. determining the ranking values of the Q search results according to the Q relevancy degrees and the sequence number value of each search result of the Q search results corresponding to the Q relevancy degrees to obtain Q ranking values;
745. and sequencing the Q search results according to the Q sequencing values to obtain the target search result set.
In a specific implementation, each search result may correspond to a browsing record of the user, and each browsing record may correspond to age information of the user, in a specific implementation, the search user may be a registered user, and when looking up each search result, the search user may generate a corresponding browsing record, and the browsing record may also record the age information of the user. Furthermore, browsing records of each search result in the P search results can be obtained to obtain P browsing records, and each browsing record corresponds to the age information of the user.
Furthermore, a browsing distribution histogram corresponding to each search result in the P search results can be determined according to the P browsing records to obtain P browsing distribution histograms, the horizontal axis of the browsing distribution histogram is age, and the vertical axis is the number of users, and the association degree between each search result in the P search results and the target age can be determined according to the P browsing distribution histograms, so as to obtain P association degrees, for example, the number of users corresponding to the target age may be obtained, and a ratio between the number of users and the total number of users may be determined, and the ratio may be used as a degree of association, for example, the number of users in a preset age range including the target age may be selected, a ratio between the number of users and the total number of users may be determined, the ratio may be used as the degree of association, for example, the preset age range includes an upper limit value and a lower limit value, and the target age may be a mean value of the upper limit value and the lower limit value.
Furthermore, the preset range can be set by a user or defaulted by a system, the relevance in the preset range can be screened out from the P relevance to obtain Q relevance, Q is a positive integer less than or equal to P, the ranking values of Q search results are determined according to the Q relevance and the sequence number value of each search result of the Q search results corresponding to the Q relevance to obtain Q ranking values, and finally the Q search results are ranked according to the sequence of the Q ranking values from large to small to obtain a target search result set.
Optionally, in the step 744, determining the ranking values of the Q search results according to the Q relevance degrees and the sequence number value of each search result of the Q search results corresponding to the Q relevance degrees, to obtain Q ranking values, which may include the following steps:
7441. determining a target adjusting factor corresponding to the association degree i according to a preset mapping relation between the association degree and the adjusting factor, wherein the association degree i is any one of the Q association degrees;
7442. determining a target sequence number value corresponding to the sequence number of the association degree i according to a mapping relation between a preset sequence number and the sequence number value, wherein the sequence number value is larger the earlier the sequence number is;
7443. and adjusting the target sequence number value according to the target adjusting factor to obtain a ranking value corresponding to the association degree i.
The mapping relationship between the preset association degree and the adjustment factor may be pre-stored, the value range of the adjustment factor may be 0-1, or the adjustment factor may also be a specific numerical value, and the mapping relationship between the preset sequence number and the sequence number value.
Specifically, taking the association degree i as an example, the association degree i is any association degree in Q association degrees, and a target adjustment factor corresponding to the association degree i may be determined according to a mapping relationship between a preset association degree and an adjustment factor, and then a target sequence number value corresponding to a sequence number of the association degree i may be determined according to a mapping relationship between a preset sequence number and a sequence number value, where the sequence number is larger the farther forward, and conversely, the sequence number is smaller the farther backward, and further, the target sequence number value may be adjusted according to the target adjustment factor to obtain a ranking value corresponding to the association degree i, which is as follows:
rank value of degree of association i = (1 + target adjustment factor) × target number value
Furthermore, the sequence number value corresponding to the sequence number can be adjusted according to the relevance, so that the adjusted sequencing value is more strongly related to the age stage of the user, the sequencing result is more suitable for the age of the user, and the user experience is improved.
It can be seen that the voice search method described in the embodiment of the present application obtains a target voice input by a user, converts the target voice into a target text, performs feature extraction on the target text to obtain a first keyword set, obtains a target browsing record of the user within a preset time period, performs feature extraction on the target browsing record to obtain a second keyword set, determines a third keyword set according to the first keyword set and the second keyword set, and performs search according to the third keyword set to obtain a target search result set, so that the voice can be converted into a text and corresponding keywords can be extracted, keywords in the browsing record can be extracted, and then the keywords in the two are integrated, the integrated keywords include a user search appeal expressed by the voice of the user, and also include keywords which may be inspired by the user in the browsing record, such third keyword has a strong user intention, in the searching process, the searching result required by the user can be obtained more favorably.
Referring to fig. 4, fig. 4 is a schematic flow chart of a voice search method according to an embodiment of the present application, where as shown in the figure, the voice search method includes:
401. and acquiring target voice input by a user.
402. And preprocessing the target voice to obtain the preprocessed target voice.
403. And converting the preprocessed target voice into a target text.
404. And extracting the characteristics of the target text to obtain a first keyword set.
405. And acquiring a target browsing record of the user in a preset time period.
406. Extracting features of the target browsing record to obtain a second keyword set;
407. and determining a third keyword set according to the first keyword set and the second keyword set.
408. And searching according to the third keyword set to obtain a target search result set.
The specific description of steps 401 to 408 may refer to the corresponding steps of the voice search method described in fig. 1, and will not be described herein again.
It can be seen that the voice search method described in the embodiment of the present application obtains a target voice input by a user, preprocesses the target voice to obtain a preprocessed target voice, converts the preprocessed target voice into a target text, performs feature extraction on the target text to obtain a first keyword set, obtains a target browsing record of the user in a preset time period, performs feature extraction on the target browsing record to obtain a second keyword set, determines a third keyword set according to the first keyword set and the second keyword set, performs search according to the third keyword set to obtain a target search result set, so that the voice can be converted into a text and corresponding keywords are extracted, the keywords in the browsing record are integrated, and the integrated keywords include a user search appeal expressed by the voice of the user, the keywords which are possibly inspired by the user in the browsing record are also covered, and the third keyword has stronger user intention, so that the searching result required by the user can be obtained more favorably in the searching process.
In accordance with the foregoing embodiments, please refer to fig. 5, fig. 5 is a schematic structural diagram of a server according to an embodiment of the present application, and as shown in the drawing, the server includes a processor, a memory, a communication interface, and one or more programs, the one or more programs are stored in the memory and configured to be executed by the processor, and in an embodiment of the present application, the programs include instructions for performing the following steps:
acquiring target voice input by a user;
converting the target voice into a target text;
extracting features of the target text to obtain a first keyword set;
acquiring a target browsing record of the user in a preset time period;
extracting features of the target browsing record to obtain a second keyword set;
determining a third keyword set according to the first keyword set and the second keyword set;
and searching according to the third keyword set to obtain a target search result set.
Optionally, in the aspect of determining a third keyword set according to the first keyword set and the second keyword set, the program includes instructions for performing the following steps:
selecting keywords related to the keywords in the first keyword set from the second keyword set to obtain at least one keyword;
merging the second keyword set and the at least one keyword to obtain a fourth keyword set;
and selecting a preset number of keywords from the fourth keyword set as the third keyword set.
Optionally, in the aspect of obtaining a target search result set by performing a search according to the third key word set, the program includes instructions for performing the following steps:
searching according to the third keyword set to obtain an initial search result set;
selecting P search results which are ranked at the top from the initial search result set, wherein P is an integer larger than 1;
acquiring the age of the user to obtain a target age;
and screening and sequencing the P search results according to the target age to obtain the target search result set.
Optionally, in the aspect that the P search results are filtered and ranked according to the target age to obtain the target search result set, the program includes instructions for performing the following steps:
acquiring browsing records of each search result in the P search results to obtain P browsing records, wherein each browsing record corresponds to the age information of the user;
determining a browsing distribution histogram corresponding to each search result in the P search results according to the P browsing records to obtain P browsing distribution histograms, wherein the horizontal axis of the browsing distribution histogram is age, and the vertical axis of the browsing distribution histogram is user number;
determining the association degree between each search result in the P search results and the target age according to the P browsing distribution histograms to obtain P association degrees;
screening out the relevance in a preset range from the P relevance to obtain Q relevance, wherein Q is a positive integer less than or equal to P;
determining the ranking values of the Q search results according to the Q relevancy degrees and the sequence number value of each search result of the Q search results corresponding to the Q relevancy degrees to obtain Q ranking values;
and sequencing the Q search results according to the Q sequencing values to obtain the target search result set.
Optionally, in the aspect that the ranking values of the Q search results are determined according to the Q relevance degrees and the sequence number value of each search result of the Q search results corresponding to the Q relevance degrees, so as to obtain the Q ranking values, the program includes instructions for executing the following steps:
determining a target adjusting factor corresponding to the association degree i according to a preset mapping relation between the association degree and the adjusting factor, wherein the association degree i is any one of the Q association degrees;
determining a target sequence number value corresponding to the sequence number of the association degree i according to a mapping relation between a preset sequence number and the sequence number value, wherein the sequence number value is larger the earlier the sequence number is;
and adjusting the target sequence number value according to the target adjusting factor to obtain a ranking value corresponding to the association degree i.
Optionally, in the aspect of converting the target speech into the target text, the program includes instructions for performing the following steps:
identifying the target voice to obtain a target language type;
determining a target voice recognition model corresponding to the target voice type according to a mapping relation between a preset language type and a voice recognition model;
and converting the target voice into a target text through the target voice recognition model.
It can be seen that, in the server described in this embodiment of the present application, the target voice input by the user is obtained, the target voice is converted into the target text, the feature extraction is performed on the target text to obtain the first keyword set, the target browsing record of the user in a preset time period is obtained, the feature extraction is performed on the target browsing record to obtain the second keyword set, the third keyword set is determined according to the first keyword set and the second keyword set, and the search is performed according to the third keyword set to obtain the target search result set, so that the voice can be converted into the text, corresponding keywords can be extracted, keywords in the browsing record can be extracted, and then the keywords in the two are integrated, the integrated keywords include the search appeal of the user expressed by the voice of the user, and also include keywords that the user in the browsing record may be inspired by the user, such third keyword has a strong user intention, in the searching process, the searching result required by the user can be obtained more favorably.
Fig. 6 is a block diagram showing functional units of a speech search apparatus 600 according to an embodiment of the present application. The voice searching device 600, the device 600 comprises: a first acquisition unit 601, a translation unit 602, a first extraction unit 603, a second acquisition unit 604, a second extraction unit 605, a determination unit 606, and a search unit 607, wherein,
the first obtaining unit 601 is configured to obtain a target voice input by a user;
the converting unit 602 is configured to convert the target speech into a target text;
the first extraction unit 603 is configured to perform feature extraction on the target text to obtain a first keyword set;
the second obtaining unit 604 is configured to obtain a target browsing record of the user in a preset time period;
the second extraction unit 605 is configured to perform feature extraction on the target browsing record to obtain a second keyword set;
the determining unit 606 is configured to determine a third keyword set according to the first keyword set and the second keyword set;
the searching unit 607 is configured to perform a search according to the third keyword set to obtain a target search result set.
Optionally, in the aspect of determining a third keyword set according to the first keyword set and the second keyword set, the determining unit 606 is specifically configured to:
selecting keywords related to the keywords in the first keyword set from the second keyword set to obtain at least one keyword;
merging the second keyword set and the at least one keyword to obtain a fourth keyword set;
and selecting a preset number of keywords from the fourth keyword set as the third keyword set.
Optionally, in the aspect of obtaining a target search result set by performing the search according to the third key word set, the search unit 607 is specifically configured to:
searching according to the third keyword set to obtain an initial search result set;
selecting P search results which are ranked at the top from the initial search result set, wherein P is an integer larger than 1;
acquiring the age of the user to obtain a target age;
and screening and sequencing the P search results according to the target age to obtain the target search result set.
Optionally, in the aspect that the P search results are filtered and ranked according to the target age to obtain the target search result set, the search unit 607 is specifically configured to:
acquiring browsing records of each search result in the P search results to obtain P browsing records, wherein each browsing record corresponds to the age information of the user;
determining a browsing distribution histogram corresponding to each search result in the P search results according to the P browsing records to obtain P browsing distribution histograms, wherein the horizontal axis of the browsing distribution histogram is age, and the vertical axis of the browsing distribution histogram is user number;
determining the association degree between each search result in the P search results and the target age according to the P browsing distribution histograms to obtain P association degrees;
screening out the relevance in a preset range from the P relevance to obtain Q relevance, wherein Q is a positive integer less than or equal to P;
determining the ranking values of the Q search results according to the Q relevancy degrees and the sequence number value of each search result of the Q search results corresponding to the Q relevancy degrees to obtain Q ranking values;
and sequencing the Q search results according to the Q sequencing values to obtain the target search result set.
Optionally, in the aspect that the ranking values of the Q search results are determined according to the Q relevance degrees and the sequence number value of each search result of the Q search results corresponding to the Q relevance degrees, so as to obtain the Q ranking values, the search unit 607 is specifically configured to:
determining a target adjusting factor corresponding to the association degree i according to a preset mapping relation between the association degree and the adjusting factor, wherein the association degree i is any one of the Q association degrees;
determining a target sequence number value corresponding to the sequence number of the association degree i according to a mapping relation between a preset sequence number and the sequence number value, wherein the sequence number value is larger the earlier the sequence number is;
and adjusting the target sequence number value according to the target adjusting factor to obtain a ranking value corresponding to the association degree i.
Optionally, in terms of converting the target speech into the target text, the conversion unit 602 is specifically configured to:
identifying the target voice to obtain a target language type;
determining a target voice recognition model corresponding to the target voice type according to a mapping relation between a preset language type and a voice recognition model;
and converting the target voice into a target text through the target voice recognition model.
It can be seen that the voice search apparatus described in the embodiment of the present application obtains a target voice input by a user, converts the target voice into a target text, performs feature extraction on the target text to obtain a first keyword set, obtains a target browsing record of the user within a preset time period, performs feature extraction on the target browsing record to obtain a second keyword set, determines a third keyword set according to the first keyword set and the second keyword set, and performs search according to the third keyword set to obtain a target search result set, so that the voice can be converted into a text and corresponding keywords can be extracted, keywords in the browsing record can be extracted, and then the keywords in the two are integrated, the integrated keywords include a user search appeal expressed by the voice of the user, and also include keywords which may be inspired by the user in the browsing record, such third keyword has a strong user intention, in the searching process, the searching result required by the user can be obtained more favorably.
It can be understood that the functions of each program module of the speech search apparatus in this embodiment may be specifically implemented according to the method in the foregoing method embodiment, and the specific implementation process may refer to the related description of the foregoing method embodiment, which is not described herein again.
Embodiments of the present application also provide a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any one of the methods as described in the above method embodiments.
Embodiments of the present application also provide a computer program product comprising a non-transitory computer readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the methods as described in the above method embodiments. The computer program product may be a software installation package.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present application is not limited by the order of acts described, as some steps may occur in other orders or concurrently depending on the application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required in this application.
In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, the above-described division of the units is only one type of division of logical functions, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of some interfaces, devices or units, and may be an electric or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit may be stored in a computer readable memory if it is implemented in the form of a software functional unit and sold or used as a stand-alone product. Based on such understanding, the technical solution of the present application may be substantially implemented or a part of or all or part of the technical solution contributing to the prior art may be embodied in the form of a software product stored in a memory, and including several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the above-mentioned method of the embodiments of the present application. And the aforementioned memory comprises: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable memory, which may include: flash Memory disks, Read-Only memories (ROMs), Random Access Memories (RAMs), magnetic or optical disks, and the like.
The foregoing detailed description of the embodiments of the present application has been presented to illustrate the principles and implementations of the present application, and the above description of the embodiments is only provided to help understand the method and the core concept of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (10)
1. A method for voice searching, the method comprising:
acquiring target voice input by a user;
converting the target voice into a target text;
extracting features of the target text to obtain a first keyword set;
acquiring a target browsing record of the user in a preset time period;
extracting features of the target browsing record to obtain a second keyword set;
determining a third keyword set according to the first keyword set and the second keyword set;
and searching according to the third keyword set to obtain a target search result set.
2. The method of claim 1, wherein determining a third set of keywords from the first set of keywords and the second set of keywords comprises:
selecting keywords related to the keywords in the first keyword set from the second keyword set to obtain at least one keyword;
merging the second keyword set and the at least one keyword to obtain a fourth keyword set;
and selecting a preset number of keywords from the fourth keyword set as the third keyword set.
3. The method of claim 1 or 2, wherein the searching according to the third keyword set to obtain a target search result set comprises:
searching according to the third keyword set to obtain an initial search result set;
selecting P search results which are ranked at the top from the initial search result set, wherein P is an integer larger than 1;
acquiring the age of the user to obtain a target age;
and screening and sequencing the P search results according to the target age to obtain the target search result set.
4. The method of claim 3, wherein said filtering and ranking said P search results according to said target age to obtain said target search result set comprises:
acquiring browsing records of each search result in the P search results to obtain P browsing records, wherein each browsing record corresponds to the age information of the user;
determining a browsing distribution histogram corresponding to each search result in the P search results according to the P browsing records to obtain P browsing distribution histograms, wherein the horizontal axis of the browsing distribution histogram is age, and the vertical axis of the browsing distribution histogram is user number;
determining the association degree between each search result in the P search results and the target age according to the P browsing distribution histograms to obtain P association degrees;
screening out the relevance in a preset range from the P relevance to obtain Q relevance, wherein Q is a positive integer less than or equal to P;
determining the ranking values of the Q search results according to the Q relevancy degrees and the sequence number value of each search result of the Q search results corresponding to the Q relevancy degrees to obtain Q ranking values;
and sequencing the Q search results according to the Q sequencing values to obtain the target search result set.
5. The method according to claim 4, wherein determining the ranking values of the Q search results according to the Q relevancy degrees and the sequence number value of each search result of the Q search results corresponding to the Q relevancy degrees to obtain Q ranking values comprises:
determining a target adjusting factor corresponding to the association degree i according to a preset mapping relation between the association degree and the adjusting factor, wherein the association degree i is any one of the Q association degrees;
determining a target sequence number value corresponding to the sequence number of the association degree i according to a mapping relation between a preset sequence number and the sequence number value, wherein the sequence number value is larger the earlier the sequence number is;
and adjusting the target sequence number value according to the target adjusting factor to obtain a ranking value corresponding to the association degree i.
6. The method of claim 1 or 2, wherein the converting the target speech into the target text comprises:
identifying the target voice to obtain a target language type;
determining a target voice recognition model corresponding to the target voice type according to a mapping relation between a preset language type and a voice recognition model;
and converting the target voice into a target text through the target voice recognition model.
7. A speech searching apparatus, characterized in that the apparatus comprises: a first obtaining unit, a converting unit, a first extracting unit, a second obtaining unit, a second extracting unit, a determining unit and a searching unit, wherein,
the first acquisition unit is used for acquiring target voice input by a user;
the conversion unit is used for converting the target voice into a target text;
the first extraction unit is used for extracting the characteristics of the target text to obtain a first keyword set;
the second obtaining unit is used for obtaining a target browsing record of the user in a preset time period;
the second extraction unit is used for extracting features of the target browsing record to obtain a second keyword set;
the determining unit is used for determining a third keyword set according to the first keyword set and the second keyword set;
and the searching unit is used for searching according to the third key word set to obtain a target searching result set.
8. The apparatus according to claim 7, wherein, in said determining a third keyword set from the first keyword set and the second keyword set, the determining unit is specifically configured to:
selecting keywords related to the keywords in the first keyword set from the second keyword set to obtain at least one keyword;
merging the second keyword set and the at least one keyword to obtain a fourth keyword set;
and selecting a preset number of keywords from the fourth keyword set as the third keyword set.
9. A server, comprising a processor, a memory for storing one or more programs and configured to be executed by the processor, the programs comprising instructions for performing the steps in the method of any of claims 1-6.
10. A computer-readable storage medium, characterized in that a computer program for electronic data exchange is stored, wherein the computer program causes a computer to perform the method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110871873.XA CN113312541B (en) | 2021-07-30 | 2021-07-30 | Voice search method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110871873.XA CN113312541B (en) | 2021-07-30 | 2021-07-30 | Voice search method, device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113312541A true CN113312541A (en) | 2021-08-27 |
| CN113312541B CN113312541B (en) | 2021-10-26 |
Family
ID=77382198
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110871873.XA Active CN113312541B (en) | 2021-07-30 | 2021-07-30 | Voice search method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113312541B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070061332A1 (en) * | 2005-09-14 | 2007-03-15 | Jorey Ramer | User history influenced search results |
| CN102521251A (en) * | 2011-11-16 | 2012-06-27 | 北京搜狗科技发展有限公司 | Method for directly realizing personalized search, device for realizing method, and search server |
| US20150154472A1 (en) * | 2013-11-29 | 2015-06-04 | Canon Kabushiki Kaisha | Search apparatus, search method, and storage medium |
| CN109800348A (en) * | 2018-12-12 | 2019-05-24 | 平安科技（深圳）有限公司 | Search for information display method, device, storage medium and server |
| US20190332612A1 (en) * | 2018-04-27 | 2019-10-31 | Branch Metrics, Inc. | Search keyword generation |
| CN111198936A (en) * | 2018-11-20 | 2020-05-26 | 北京嘀嘀无限科技发展有限公司 | Voice search method and device, electronic equipment and storage medium |
| CN111625680A (en) * | 2020-05-15 | 2020-09-04 | 青岛聚看云科技有限公司 | Method and device for determining search result |
| CN111782956A (en) * | 2020-07-08 | 2020-10-16 | 重庆帮企科技集团有限公司 | Search method based on user behavior and keyword classification |
-
2021
- 2021-07-30 CN CN202110871873.XA patent/CN113312541B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070061332A1 (en) * | 2005-09-14 | 2007-03-15 | Jorey Ramer | User history influenced search results |
| CN102521251A (en) * | 2011-11-16 | 2012-06-27 | 北京搜狗科技发展有限公司 | Method for directly realizing personalized search, device for realizing method, and search server |
| US20150154472A1 (en) * | 2013-11-29 | 2015-06-04 | Canon Kabushiki Kaisha | Search apparatus, search method, and storage medium |
| US20190332612A1 (en) * | 2018-04-27 | 2019-10-31 | Branch Metrics, Inc. | Search keyword generation |
| CN111198936A (en) * | 2018-11-20 | 2020-05-26 | 北京嘀嘀无限科技发展有限公司 | Voice search method and device, electronic equipment and storage medium |
| CN109800348A (en) * | 2018-12-12 | 2019-05-24 | 平安科技（深圳）有限公司 | Search for information display method, device, storage medium and server |
| CN111625680A (en) * | 2020-05-15 | 2020-09-04 | 青岛聚看云科技有限公司 | Method and device for determining search result |
| CN111782956A (en) * | 2020-07-08 | 2020-10-16 | 重庆帮企科技集团有限公司 | Search method based on user behavior and keyword classification |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113312541B (en) | 2021-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108287858B (en) | Semantic extraction method and device for natural language | |
| US11194965B2 (en) | Keyword extraction method and apparatus, storage medium, and electronic apparatus | |
| CN108829822B (en) | Media content recommendation method and device, storage medium and electronic device | |
| KR102455616B1 (en) | Theme classification method based on multimodality, device, apparatus, and storage medium | |
| CN108829893B (en) | Method and device for determining video label, storage medium and terminal equipment | |
| KR102222317B1 (en) | Speech recognition method, electronic device, and computer storage medium | |
| CN107797984B (en) | Intelligent interaction method, equipment and storage medium | |
| CN110444198B (en) | Retrieval method, retrieval device, computer equipment and storage medium | |
| CN110597952A (en) | Information processing method, server, and computer storage medium | |
| KR20180078318A (en) | Methods and Apparatus for Determining the Agents | |
| CN111326140B (en) | Speech recognition result discriminating method, correcting method, device, equipment and storage medium | |
| CN113094552A (en) | Video template searching method and device, server and readable storage medium | |
| CN110072140B (en) | Video information prompting method, device, equipment and storage medium | |
| US11361759B2 (en) | Methods and systems for automatic generation and convergence of keywords and/or keyphrases from a media | |
| US9454568B2 (en) | Method, apparatus and computer storage medium for acquiring hot content | |
| CN116882372A (en) | Text generation method, device, electronic equipment and storage medium | |
| CN116150306A (en) | Training method of question-answering robot, question-answering method and device | |
| CN110738061A (en) | Ancient poetry generation method, device and equipment and storage medium | |
| CN113312541B (en) | Voice search method, device and storage medium | |
| CN110797013A (en) | Live broadcast entrance display method of voice live broadcast room, related equipment and storage medium | |
| CN109947947B (en) | Text classification method and device and computer readable storage medium | |
| CN112735395A (en) | Voice recognition method, electronic equipment and storage device | |
| CN113836346B (en) | Method, device, computing equipment and storage medium for generating abstract for audio file | |
| TWI767197B (en) | Method and server for providing interactive voice tutorial | |
| CN116913278B (en) | Voice processing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |