Voice search method, search device, readable storage medium and electronic equipment

Abstract

The application provides a voice search method, a search device, a readable storage medium and an electronic equipment. The voice search method comprises the following steps: detecting a voice keyword to identify a voice search intention; after determining the voice search intention, detecting voice on-screen content and performing a search; wherein the search operation on the current voice on-screen content is performed once every multiple preset rules of text of the voice on-screen content, so as to continuously update the previous search result; and the search result is displayed on the screen. The voice search scheme of saying while searching is provided. Compared with the technical scheme in the prior art that waits for the voice input to be completed and then starts searching, the voice search scheme has a faster search speed, and can effectively improve the user experience.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of voice search, in particular to a voice search method, a search device, a readable storage medium and an electronic device. BACKGROUND

[0002] Generally, the voice search function needs the user to finish speaking the to-be-searched keyword phrase before starting the search, and there is an interval between the voice on-screen and the final determination of the end of the voice, so starting the search after determining the end of the voice will make the search speed relatively slow.

[0003] In a search application, a name search (including a surrounding area) and a sug search (i.e. a place input prompt retrieval, also known as a POI hotword suggestion retrieval, an online suggestion retrieval, a Suggestion POI search, which refers to online suggestion word query according to a keyword) are usually adopted, and the search based on voice recognition is generally one-time, needs to obtain sufficient information (such as poi details), and the speed of the name search is slower than that of the sug search, and is usually twice as slow.

[0004] In view of this, the industry urgently needs to propose a new voice search technology to solve the problem of slow speed of the existing voice search technology. SUMMARY

[0005] In view of the above-mentioned defects of the prior art, the purpose of the present application is to provide a voice search method, a search device, a readable storage medium and an electronic device for solving the above problems.

[0006] To achieve the above-mentioned purpose and other related purposes, the present application provides a voice search method, comprising: detecting a voice keyword to identify a voice search intention; after determining the voice search intention, detecting voice on-screen content and performing a search; wherein the voice on-screen content performs a search operation on the current voice on-screen content every more than a preset rule of text to constantly update the previous search result; and displaying the search result on the screen.

[0007] In an embodiment of the present application, the method further comprises: when the voice on-screen content of the first time and the intermediate times is searched, a sug search is adopted; when the voice on-screen content of the last time is searched, a sug search and a name / surrounding area search are respectively adopted, and after the sug search result and the name / surrounding area search result are fused, a final search result is formed.

[0008] In an embodiment of the present application, the method further comprises: displaying the search result in the form of a list on the screen, and marking the corresponding geographical position in an electronic map.

[0009] In an embodiment of the present application, for the several intermediate times of the voice on-screen content, the method further comprises: searching for each of the intermediate times of the voice on-screen content respectively; sorting the search results of the first time of the voice on-screen content, the search results of each of the intermediate times of the voice on-screen content, and the search results of the last time of the voice on-screen content according to time; and forming the final voice search results after removing the false results of each of the search results.

[0010] To achieve the above object and other related objects, the present application provides a voice search device, comprising: an intention recognition module, configured to detect a voice keyword to recognize a voice search intention; a content search module, configured to detect voice on-screen content and perform a search after the intention recognition module determines the voice search intention; wherein the voice on-screen content performs a search operation on the current voice on-screen content every multiple preset rules of text to constantly update the previous search results; and a result display module, configured to display the search results on a screen.

[0011] In an embodiment of the present application, the content search module is further configured to: when searching for the first time and the intermediate times of the voice on-screen content, adopt sug search; when searching for the last time of the voice on-screen content, adopt sug search and name / surrounding search respectively, and form the final search results after fusing the sug search results and the name / surrounding search results.

[0012] In an embodiment of the present application, the result display module is further configured to: display the search results in the form of a list on the screen, and mark the corresponding geographical positions in an electronic map.

[0013] In an embodiment of the present application, for the several intermediate times of the voice on-screen content, the content search module is further configured to: search for each of the intermediate times of the voice on-screen content respectively; sort the search results of the first time of the voice on-screen content, the search results of each of the intermediate times of the voice on-screen content, and the search results of the last time of the voice on-screen content according to time; and form the final voice search results after removing the false results of each of the search results.

[0014] To achieve the above object and other related objects, the present application provides a computer readable storage medium, wherein a computer program is stored, and the computer program is loaded and executed by a processor to implement the voice search method.

[0015] To achieve the above object and other related objects, the present application provides an electronic device, comprising: a processor and a memory; wherein the memory is configured to store a computer program; and the processor is configured to load and execute the computer program to enable the electronic device to perform the voice search method.

[0016] As described above, the voice search method, the search device, the readable storage medium and the electronic device of the present application detect a voice keyword to identify a voice search intention; after determining the voice search intention, a voice on-screen content is detected and searched; wherein the voice on-screen content performs a search operation on the current voice on-screen content every multiple of a predetermined rule of text to constantly update the previous search result; and the search result is displayed on the screen. The present application provides a voice search scheme of saying while searching, compared with the technical scheme of waiting for the voice input to end before starting to search in the prior art, the search speed is faster, and the user experience can be effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A flow chart of a voice search method in an embodiment of the present application is shown.

[0018] Figures 2a-2b An interactive interface diagram for implementing a voice search method in an embodiment of the present application is shown.

[0019] Figure 3 A module diagram of a voice search device in an embodiment of the present application is shown.

[0020] Figure 4 A structure diagram of an electronic device in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0021] The embodiments of the present application will be described in detail hereinafter with specific reference to the drawings. Other advantages and effects of the present application can be easily understood by those skilled in the art from the content disclosed in the specification. The present application can also be implemented or applied by different specific embodiments, and the details in the specification can be modified or changed based on different views and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0022] It should be noted that the diagrams provided in the following embodiments only illustrate the basic concept of the present application in a schematic manner, and only the components related to the present application are shown in the diagrams, not the number, shape and size of the components when actually implemented. The shape, number and proportion of each component when actually implemented can be arbitrarily changed, and the layout pattern of the components can also be more complex.

[0023] Currently, the voice search process needs to wait for the voice to end and will be performed after the ASR recognition callback is completed, and the longer the search keyword is, the longer the waiting time is, and the user experience is poor. The voice search scheme of speaking while searching is provided, and the search can be started as long as the recognized text is obtained, and the result needed by the user may be searched out before the user finishes speaking, for example, the voice "I want to go to Hongqiao Railway Station", and the application may display the Hongqiao Railway Station in the result list after recognizing the user voice "I want to go to Hongqiao", which greatly improves the voice search efficiency and improves the user experience.

[0024] As shown in Figure 1 The voice search method of the application mainly includes the following steps:

[0025] S11: detecting a voice keyword to recognize a voice search intention;

[0026] Specifically, the voice keyword can be "navigate to", "I want to go", "nearby", etc., and the voice search intention can be recognized by detecting these voice keywords in the user voice, so as to start the voice search mode of speaking while searching.

[0027] S12: after determining the voice search intention, detecting voice on-screen content and performing search; wherein the search operation on the current voice on-screen content is performed once every more than a preset rule of text, so as to constantly update the previous search result; wherein the preset rule of text can be a fixed number of texts.

[0028] Specifically, an input box is displayed on the screen, and the user voice content is displayed in the input box in real time, which is called "voice on screen", for example, when the user speaks "Hong", the input box displays "Hong", and when the user speaks "Qiao" immediately, the input box displays "Hongqiao", and so on. The content in the input box is detected to obtain the object to be searched, the voice keyword is not on screen, that is, it is not displayed in the input box, and the on-screen content is all used as the object to be searched, for example, the user voice is "I want to go to Hongqiao", and the content in the input box is "Hongqiao", and the search operation is performed based on "Hongqiao".

[0029] Preferably, the application performs a search operation once for each additional character in the input box. Taking the complete voice of the user "Hongqiao Railway Station" as an example, when "Hong" is displayed in the input box, a search for "Hong" is performed and the search result is obtained; when "Hongqiao" is displayed in the input box, a search for "Hongqiao" is performed and the search result is obtained; when "Hongqiao Fire" is displayed in the input box, a search for "Hongqiao Fire" is performed and the search result is obtained; when "Hongqiao Railway" is displayed in the input box, a search for "Hongqiao Railway" is performed and the search result is obtained; when "Hongqiao Railway Station" is displayed in the input box, a search for "Hongqiao Railway Station" is performed and the search result is obtained. With the continuous search, the newly generated search result updates the previous search result.

[0030] S13: displaying the search result on the screen.

[0031] Specifically, the search result can be displayed on the screen in the form of a list, and the corresponding geographical position can be marked on the electronic map, wherein each search result in the list can be sorted according to the distance from the current position or a preset hot word, and the search result with a high recommendation degree is arranged at the top.

[0032] In an embodiment, the step S12 uses sug search to implement the search for the first and intermediate search contents, and uses name / surrounding search (specifically, name search or surrounding search depending on the type of the search object, such as a certain intersection, a certain address, a certain building, a certain restaurant, etc.) and sug search to respectively implement the search for the last search content, and after the sug search result and the name / surrounding search result are fused, the final search result is formed. Taking "Hongqiao Railway Station" as an example, "Hong" is the first search content, "Hongqiao", "Hongqiao Fire", "Hongqiao Railway" are the intermediate search contents, and "Hongqiao Railway Station" is the last search content.

[0033] It should be noted that the skilled person in the art can use the existing ASR recognition to determine whether the user voice is ended, and then identify the last search content. Since this is not the key point of the application, it will not be introduced here. After the user voice is ended, the last content is used to perform a name / surrounding search again, and the purpose is that since the content at the last time of on-screen and the content of the last name / surrounding search are the same, the content displayed after the sug search is sufficient to fill the display list, and when the result of the name / surrounding search and the content of the sug search are fused, it can be seen that the list and map display change little, and the experience is good.

[0034] In an embodiment, the step S12 further performs the following steps for the intermediate search contents:

[0035] Firstly, sug search is conducted on each intermediate speech content, i.e. multi-thread search; multi-thread synchronous search can ensure that the user can get the corresponding result display for each word during the speech process, and the process is relatively fast, and the final search result is the fused result, and the search result is relatively comprehensive;

[0036] Secondly, the sug search results of the first speech content, the sug search results of each intermediate speech content, and the search result of the last speech content are sorted by time;

[0037] Thirdly, after removing the error results of each search result, the final speech search result is formed.

[0038] It should be noted that in the actual operation process, for multiple intermediate sug searches, the search time required for the first search content may be long, and the search time required for the last search content may be short, so the result of the first search is generated later than the result of the last search, which leads to the inability to correctly use the result of the last search to update the result of the first search, resulting in confusion. Therefore, by arranging each search result according to the search order and excluding some error results, the result of the last search is used to update the result of the first search, and the final search result can be correctly obtained.

[0039] The speech search method of the present application searches while speaking, which adopts sug search in the intermediate search link, and the search speed is fast enough. However, sug search only contains basic information and the number is relatively small, and the returned result is only a few, and at most 10 returned results (generally 200 milliseconds can get the query result), therefore, in order to ensure that the final result obtained by the speech search method of the present application is sufficient, sug search and name / peripheral search are conducted on the last speech content, and the search results of the two are fused. In this way, the search result obtained by the speech search method of the present application is relatively detailed, and 20 results can be returned at a time, and the query result can be obtained in about 400 milliseconds.

[0040] In practical application, the result returned by the sug generally contains the result that the user wants, and the user does not need to wait for the name / surrounding search result to come out, that is, in practical application, the user can get the desired search result when not saying the complete sentence, and can not continue to say the latter words. Therefore, in the specific steps, the sug search is selected in the present application, which can be much faster than the speed of returning the result by searching according to the key word after the user finishes saying all the content in the prior art. In addition, the traditional voice search has a long period from the first word recognized by voice recognition (i.e., on-screen display) to the end of the voice and then to the semantic output. After the semantic callback, several seconds are spent in searching, and there is a certain interval from the end of the voice to the semantic callback. However, the interval problem is effectively avoided by the present application through the search while speaking, so that the search speed is further improved.

[0041] The implementation process of the voice search method will be described below. Figures 2a-2b The implementation process of the voice search method will be described below.

[0042] The complete voice of the user is "navigate to Zifeng Mansion". Figure 2a When the user says "navigate to Zifeng Da", "Zifeng Da" is on-screen and displayed in the address input box, and the search result of "Zifeng Da" is displayed in the list below, and the geographical position corresponding to the list result is marked on the electronic map on the right. Figure 2b When the user says "navigate to Zifeng Mansion", "Zifeng Mansion" is on-screen and displayed in the address input box, and the latest search result of "Zifeng Mansion" is displayed in the list below, and the geographical position corresponding to the list result is marked on the electronic map on the right.

[0043] All or part of the steps of the above-mentioned method embodiments can be completed by computer program related hardware. Based on such understanding, the present application also provides a computer program product including one or more computer instructions. The computer instructions can be stored in a computer readable storage medium. The computer readable storage medium can be any available medium that can be stored by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available media can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)), etc.

[0044] Referring to Figure 3 The present embodiment provides a voice search device 30. Since the technical principle of the present embodiment is similar to that of the above-mentioned method embodiment, the same technical details will not be repeated. The voice search device 30 of the present embodiment includes the following modules: an intent recognition module 31, a content search module 32, and a result display module 33.

[0045] The intent recognition module 31 is configured to perform step S11 of the foregoing method embodiment, the content search module 32 is configured to perform step S12 of the foregoing method embodiment, and the result display module 33 is configured to perform step S13 of the foregoing method embodiment.

[0046] The content search module 32 is further configured to: when searching the first-time and intermediate-time voice on-screen content, using the sug search; when searching the last-time voice on-screen content, using the sug search and the name / surrounding search respectively, and forming the final search result after fusing the sug search result and the name / surrounding search result. For the intermediate-time voice on-screen content, searching each intermediate-time voice on-screen content respectively; sorting the search result of the first-time voice on-screen content, the search result of each intermediate-time voice on-screen content, and the search result of the last-time voice on-screen content according to time; and forming the final voice search result after removing the error result of each search result.

[0047] The result display module 32 is further configured to: display the search result in the form of a list on the screen, and mark the corresponding geographical position in the electronic map.

[0048] Those skilled in the art should understand that, Figure 3 The division of each module in the embodiments is only a logical functional division, and all or part of the modules can be integrated into one or more physical entities in actual implementation. The modules can all be implemented in the form of software invoked by a processing element, or all be implemented in the form of hardware, or part of the modules be implemented in the form of software invoked by a processing element and part of the modules be implemented in the form of hardware.

[0049] With reference to Figure 4 The embodiment provides an electronic device, which can be a desktop computer, a portable computer, a smart phone, or the like. In detail, the electronic device at least includes a memory and a processor connected through a bus, wherein the memory is configured to store a computer program, and the processor is configured to execute the computer program stored in the memory to perform all or part of the steps in the foregoing method embodiments.

[0050] The system bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or only one type of bus. The communication interface is used to realize the communication between the database access device and other devices (such as a client, a read-write library and a read-only library). The memory can include a Random Access Memory (RAM), and can also include a non-volatile memory, such as at least one disk memory.

[0051] The processor mentioned above can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; can also be a Digital Signal Processing (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0052] In summary, the voice search method, the search device, the readable storage medium and the electronic equipment of the present application provide a voice search scheme of searching while speaking. Compared with the technical scheme of waiting for the voice input to end and then starting searching in the prior art, the search speed is faster, the user experience can be effectively improved, and the various shortcomings in the prior art are effectively overcome, so that the present application has a high industrial utilization value.

[0053] The above embodiments only exemplarily illustrate the principles and effects of the present application, and are not used to limit the present application. Any person skilled in the art can modify or change the above embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes completed by those skilled in the art without departing from the spirit and technical thought disclosed by the present application should be covered by the claims of the present application.

Claims

1. A voice search method, characterized in that, Includes the following steps: Detect speech keywords to identify speech search intent; After determining the voice search intent, the voice-on-screen content is detected and searched; wherein, a search operation on the current voice-on-screen content is performed once every fixed number of characters added to the voice-on-screen content, so as to continuously update the previous search results; When searching for the first and intermediate voice-on content, a SUUG search was used; in particular, a multi-threaded SUUG search was performed for each intermediate voice-on content. When searching for the last voice-activated content, both sug search and name / surrounding search are used. The search results for the first voice-activated content, the search results for each intermediate voice-activated content, and the search results for the last voice-activated content are sorted in search order. After removing erroneous results from each search result, the final search results are formed. The search results are displayed on the screen in real time.

2. The voice search method according to claim 1 further includes: The search results are displayed on the screen in a list format, and the corresponding geographical locations are marked on an electronic map.

3. A voice search device, characterized in that, Includes the following steps: The intent recognition module is used to detect voice keywords in order to identify the voice search intent; The content search module is used to detect and search the voice-on-screen content after the intent recognition module determines the voice search intent. Specifically, a search operation is performed on the current voice-on-screen content every fixed number of characters, continuously updating previous search results. When searching the first and intermediate voice-on-screen content, a SUUG search is used, with multi-threaded SUUG searches performed for each intermediate voice-on-screen content. When searching the last voice-on-screen content, both SUUG search and name / surrounding search are used. The search results for the first voice-on-screen content, the search results for each intermediate voice-on-screen content, and the search results for the last voice-on-screen content are sorted in search order. After removing erroneous results from each search result, the final search results are formed. The results display module is used to display search results on the screen in real time.

4. The voice search device according to claim 3, wherein the result display module is further configured to: display the search results on the screen in the form of a list, and mark the corresponding geographical location on an electronic map.

5. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is loaded and executed by the processor, it implements the voice search method as described in any one of claims 1 to 2.

6. An electronic device, characterized in that, include: Processor and memory; among which, The memory is used to store computer programs; The processor is used to load and execute the computer program to cause the electronic device to perform the voice search method as described in any one of claims 1 to 2.

Images