In-vehicle voice broadcast method, device and equipment and storage medium

Abstract

The invention relates to the technical field of man-machine conversation, and discloses an in-vehicle voice broadcasting method, device and equipment and a storage medium, and the method comprises the steps: carrying out the semantic analysis of each to-be-broadcasted voice instruction when a plurality of to-be-broadcasted voice instructions are received; dividing each to-be-broadcasted voice instruction according to each semantic analysis result to obtain a to-be-added category and an added category; adding each to-be-added voice instruction in the to-be-added category to a corresponding added voice instruction in the added category; and broadcasting each added voice instruction after addition. According to the invention, each to-be-broadcasted voice instruction can be divided according to the semantic analysis result of each to-be-broadcasted voice instruction, and the obtained to-be-added voice instruction is added to the added voice instruction for broadcasting. Compared with the prior art, a plurality of voice instructions to be broadcasted can be broadcasted after being added and fused, so that the broadcasting timeliness can be improved, and the user experience is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of human-computer dialogue, and in particular to an in-vehicle voice broadcasting method, device, equipment and storage medium. Background Art

[0002] At present, the car computer in the car is generally equipped with a text-to-speech (TTS) model, which converts the text information to be output by the application software (Application, APP) in the car computer into voice for broadcast, such as navigation broadcast, car fault broadcast, etc.

[0003] However, since there is only one TTS model configured in the existing car computer, when voice commands generated by multiple apps need to be broadcast at the same time, they are generally broadcast according to priority, which results in some untimely broadcasts and poor user experience.

[0004] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of the present invention is to provide a method, device, equipment and storage medium for in-vehicle voice broadcasting, aiming to solve the technical problem in the prior art that when voice commands generated by multiple APPs need to be broadcast at the same time, they are generally broadcast according to priority, which leads to some untimely broadcasts and poor user experience.

[0006] To achieve the above object, the present invention provides a method for in-vehicle voice broadcasting, the method comprising the following steps:

[0007] When receiving multiple voice instructions to be broadcasted, performing semantic analysis on each of the voice instructions to be broadcasted;

[0008] Classify the voice instructions to be broadcast according to the semantic analysis results to obtain the category to be added and the category to be added;

[0009] Adding each voice instruction to be added in the to-be-added category to the corresponding voice instruction to be added in the added category;

[0010] Each added voice command is announced.

[0011] Optionally, the step of adding each voice instruction to be added in the to-be-added category to the corresponding voice instruction to be added in the added category includes:

[0012] Determine an addable flag of each of the added voice instructions based on a semantic analysis result of each of the added voice instructions in the added category;

[0013] abbreviating each of the voice instructions to be added according to the semantic analysis result of each of the voice instructions to be added in the category to be added;

[0014] According to the addable flags, each abbreviated voice instruction to be added is added to the corresponding voice instruction to be added.

[0015] Optionally, the step of adding each abbreviated voice instruction to be added to the corresponding voice instruction to be added according to each addable flag bit includes:

[0016] Combining each abbreviated voice instruction to be added with each added voice instruction based on each addable flag to obtain a list of optional combinations;

[0017] Determine the text similarity of the added voice instructions after each combination in the optional combination list;

[0018] A target combination scheme is selected from the optional combination list based on the text similarities, and each abbreviated voice instruction to be added is added to the corresponding voice instruction to be added based on the target combination scheme.

[0019] Optionally, after the step of adding each abbreviated voice instruction to be added to the corresponding voice instruction to be added according to each addable flag, the method further includes:

[0020] Extracting features of each added voice command, and analyzing the degree of context relevance of the extracted features;

[0021] Processing each of the added voice instructions according to the analysis result;

[0022] Accordingly, the step of broadcasting each added voice instruction includes:

[0023] The processed added voice commands are announced.

[0024] Optionally, the step of processing each of the added voice instructions according to the analysis result includes:

[0025] Determine the current correlation value according to the analysis result, and judge whether the current correlation value is lower than a preset correlation threshold;

[0026] If so, each added voice instruction is supplemented or deleted.

[0027] Optionally, the step of dividing the voice instructions to be broadcast according to the semantic analysis results to obtain the categories to be added and the categories to be added includes:

[0028] Determining the importance of each of the voice instructions to be broadcast according to each semantic analysis result;

[0029] Obtaining quantity information of each voice command to be broadcast, and determining a division ratio based on the quantity information;

[0030] The voice instructions to be broadcast are divided according to the division ratio and the importance, and the category to be added and the category to be added are obtained.

[0031] Optionally, before the step of broadcasting each added voice instruction, the method further includes:

[0032] Determine a first broadcast speed and a first broadcast emphasis of the voice instruction to be added, and determine a second broadcast speed and a second broadcast emphasis of the voice instruction to be added;

[0033] Accordingly, the step of broadcasting each added voice instruction includes:

[0034] Each added voice instruction is announced according to the first announcement speed, the second announcement speed, the first announcement emphasis and the second announcement emphasis.

[0035] In addition, to achieve the above-mentioned purpose, the present invention also proposes an in-vehicle voice broadcasting device, the device comprising:

[0036] A semantic analysis module, for performing semantic analysis on each of the voice instructions to be broadcast when receiving multiple voice instructions to be broadcast;

[0037] An instruction classification module is used to classify each of the voice instructions to be broadcast according to each semantic analysis result to obtain a category to be added and a category to be added;

[0038] An instruction adding module, used for adding each voice instruction to be added in the to-be-added category to the corresponding voice instruction to be added in the added category;

[0039] The voice broadcast module is used to broadcast each added voice command.

[0040] In addition, to achieve the above-mentioned purpose, the present invention also proposes an in-vehicle voice broadcast device, which includes: a memory, a processor, and an in-vehicle voice broadcast program stored in the memory and runnable on the processor, and the in-vehicle voice broadcast program is configured to implement the steps of the in-vehicle voice broadcast method described above.

[0041] In addition, to achieve the above-mentioned purpose, the present invention also proposes a storage medium, on which an in-vehicle voice broadcast program is stored. When the in-vehicle voice broadcast program is executed by a processor, the steps of the in-vehicle voice broadcast method described above are implemented.

[0042] The present invention provides a method, device, equipment and storage medium for in-vehicle voice broadcasting. The method performs semantic analysis on each of the voice instructions to be broadcasted when receiving multiple voice instructions to be broadcasted; divides each of the voice instructions to be broadcasted according to the results of each semantic analysis to obtain a category to be added and a category to be added; adds each voice instruction to be added in the category to be added to the corresponding added voice instruction in the category to be added; and broadcasts each added voice instruction. Since the present invention can first divide each voice instruction to be broadcasted according to the results of the semantic analysis of each voice instruction to be broadcasted, and broadcasts after adding the obtained voice instruction to be added to the added voice instruction. Compared with the existing method of broadcasting according to priority when there are multiple voice instructions, the present invention can broadcast multiple voice instructions to be broadcasted after adding and fusing, thereby improving the timeliness of the broadcast and improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A schematic diagram of the structure of an in-vehicle voice broadcast device in a hardware operating environment involved in an embodiment of the present invention;

[0044] Figure 2 It is a flow chart of the first embodiment of the in-vehicle voice broadcasting method of the present invention;

[0045] Figure 3 It is a flow chart of the second embodiment of the in-vehicle voice broadcasting method of the present invention;

[0046] Figure 4 It is a flow chart of the third embodiment of the in-vehicle voice broadcasting method of the present invention;

[0047] Figure 5 This is a structural block diagram of the first embodiment of the in-vehicle voice broadcasting device of the present invention.

[0048] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0049] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0050] Reference Figure 1 , Figure 1 The present invention is a schematic diagram of the structure of an in-vehicle voice broadcast device in the hardware operating environment involved in an embodiment of the present invention.

[0051] like Figure 1As shown, the in-vehicle voice broadcasting device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0052] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the in-vehicle voice broadcasting device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0053] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and an in-vehicle voice broadcast program.

[0054] exist Figure 1 In the in-vehicle voice broadcast device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the in-vehicle voice broadcast device of the present invention can be set in the in-vehicle voice broadcast device, and the in-vehicle voice broadcast device calls the in-vehicle voice broadcast program stored in the memory 1005 through the processor 1001, and executes the in-vehicle voice broadcast method provided by the embodiment of the present invention.

[0055] It should be noted that, at present, the car computer in the car is generally equipped with a text-to-speech (TTS) model, which converts the text information to be output by the application software (Application, APP) in the car computer into voice for broadcast, such as navigation broadcast, car fault broadcast, etc.

[0056] However, since there is only one TTS model configured in the existing car computer, when voice commands generated by multiple apps need to be broadcast at the same time, they are generally broadcast according to priority, which results in some untimely broadcasts and poor user experience.

[0057] Therefore, in order to solve the above defects, when receiving multiple voice commands to be broadcast, this embodiment can first perform semantic analysis on each voice command to be broadcast, divide each voice command to be broadcast according to the obtained semantic analysis results, obtain the category to be added and the category to be added, and then add each voice command to be added in the category to be added to the added voice command in the category to be added for broadcast. Compared with the existing broadcasting according to priority when there are multiple voice commands, this embodiment can broadcast multiple voice commands to be broadcast after adding and merging, thereby improving the timeliness of the broadcast and improving the user experience.

[0058] For ease of understanding, refer to Figures 2 to 5 The embodiments of the present invention and the following embodiments are described in detail.

[0059] Reference Figure 2 , Figure 2 Schematic diagram of the flow chart of the first embodiment of the in-vehicle voice broadcast method of the present invention.

[0060] In this embodiment, the in-vehicle voice broadcast method includes the following steps:

[0061] Step S10: when receiving a plurality of voice instructions to be broadcasted, performing semantic analysis on each of the voice instructions to be broadcasted.

[0062] It is understandable that the method of this embodiment can be applied in the scenario of voice broadcasting by the vehicle computer. The execution subject of the method of this embodiment can be a computing service device with in-vehicle voice broadcasting, network communication and program running functions, such as a vehicle computer, etc., and can also be other electronic devices that realize the same or similar functions. The following is an explanation of this embodiment and the following embodiments using the above-mentioned in-vehicle voice broadcasting device (hereinafter referred to as the device).

[0063] It should be understood that the above-mentioned voice instructions to be broadcast can be instructions used by the above-mentioned device for broadcasting. In the present embodiment, different APPs may be installed in the above-mentioned device, such as a navigation APP, a music APP, etc. The voice assistant in the above-mentioned APP may generate the above-mentioned voice instructions to be broadcast and broadcast them through the above-mentioned device. At the same time, a car computer assistant may also be set in a general car computer, and the above-mentioned car computer assistant may also generate the above-mentioned voice instructions to be broadcast for broadcasting.

[0064] In a specific implementation, when different voice assistants generate voice instructions to be broadcast at the same time, the above-mentioned device can perform semantic analysis on the received multiple voice instructions to be broadcast, wherein the semantic analysis can be to analyze the words and grammar in each voice instruction to be broadcast to obtain the intention of the voice instruction to be broadcast. The specific semantic analysis process is not limited in this embodiment.

[0065] Step S20: Divide the voice instructions to be broadcast according to the semantic analysis results to obtain the categories to be added and the categories to be added.

[0066] After performing semantic analysis on each voice instruction to be broadcast, the above-mentioned device can classify each voice instruction to be broadcast according to the semantic analysis result, and divide multiple voice instructions to be broadcast into a to-be-added category and an added category; wherein the above-mentioned to-be-added category can be the adding party that adds the voice instruction to be broadcast to another voice instruction to be broadcast, and the above-mentioned added category can be the added party that adds the voice instruction to be broadcast to another voice instruction to be broadcast.

[0067] Exemplarily, if the above-mentioned device receives two voice commands to be broadcast, respectively recorded as command A and command B, after determining the semantic analysis results of command A and command B, the above-mentioned device determines that command B can be added to command A for fusion, then command B can be classified as a category to be added, and command A can be classified as a category to be added.

[0068] It should be emphasized that, as an implementation method, the above-mentioned device can determine whether there are symbols such as commas and semicolons that separate sentences in the voice instructions to be broadcast. If so, the voice instructions to be broadcast can be classified as the added category; if not, the voice instructions to be broadcast can be classified as the category to be added; for example, if the above-mentioned device receives the voice instructions to be broadcast "There is a traffic violation photo taken two hundred meters ahead, please fasten your seat belts" and "There is one kilometer left to the destination" at the same time, because there is a comma separating the sentences in the voice instructions to be broadcast "There is a traffic violation photo taken two hundred meters ahead, please fasten your seat belts", the voice instruction to be broadcast "There is a traffic violation photo taken two hundred meters ahead, please fasten your seat belts" can be classified as the added category, and the voice instruction to be broadcast "One kilometer left to the destination" can be classified as the category to be added.

[0069] As another implementation, the above device can also divide the voice instructions to be broadcast according to the length of the text, and can divide the text with more words into the added category and the text with fewer words into the category to be added.

[0070] As another implementation method, since the above-mentioned device can subsequently abbreviate the voice instruction to be broadcast, the above-mentioned device can determine the number of words in the abbreviated instruction in the voice instruction to be broadcast, and classify the abbreviated instruction with fewer words as the category to be added, and classify the abbreviated instruction with more words as the category to be added; illustratively, continuing based on the above example, if the above-mentioned voice instruction to be broadcast "There is a traffic violation photo shoot two hundred meters ahead, please fasten your seat belts" can be abbreviated to "Traffic violation photo shoot ahead, please fasten your seat belts", the voice instruction to be broadcast "One kilometer remaining to the destination" can be abbreviated to "One kilometer remaining", since "One kilometer remaining" has fewer words than "Traffic violation photo shoot ahead, please fasten your seat belts", the voice instruction to be broadcast "There is a traffic violation photo shoot two hundred meters ahead, please fasten your seat belts" can be classified as the category to be added, and the voice instruction to be broadcast "One kilometer remaining to the destination" can be classified as the category to be added.

[0071] Furthermore, considering that the contents of some of the voice instructions to be broadcast are relatively urgent and important, it is not convenient to abbreviate them too much. In this embodiment, the above step S20 includes:

[0072] Step S21: determining the importance of each voice instruction to be broadcast according to each semantic analysis result.

[0073] It should be noted that, in the present embodiment, the above-mentioned importance levels can be divided into three levels: "important", "general" and "unimportant", and the above-mentioned device can determine the importance level of each voice instruction to be broadcast according to the semantic analysis result.

[0074] Specifically, when the above-mentioned voice instructions to be broadcast involve laws and regulations, the importance of the voice instructions to be broadcast can be set to "important", for example, the above-mentioned "Traffic violation photography ahead, please fasten your seat belts" involves laws and regulations, it can be set to "important"; when the above-mentioned voice instructions to be broadcast involve mileage reminders, traffic status reminders, etc., which play a partial role for users, the importance of the voice instructions to be broadcast can be set to "general", for example, the above-mentioned "one kilometer remaining to destination" can be set to "general"; when the above-mentioned voice instructions to be broadcast involve route adjustments, facility reminders, etc., which play a small role for users, the importance of the voice instructions to be broadcast can be set to "unimportant", for example, "there is a gas station ahead" can be set to "unimportant".

[0075] Of course, the specific importance determination rules can be set according to actual conditions, and this embodiment does not limit this.

[0076] Step S22: Obtain quantity information of each voice command to be broadcast, and determine the division ratio based on the quantity information.

[0077] It is understandable that the above quantity information can be the number of voice instructions to be broadcasted currently received, and the above division ratio can be the ratio of the number divided into the category to be added and the category to be added. Since the present embodiment only sets the category to be added and the category to be added, when there are two voice instructions to be broadcasted, the division ratio can be set to 1:1, thereby ensuring that the two voice instructions to be broadcasted can be broadcasted at the same time. When there are more than two (generally only three at most), one of the voice instructions to be broadcasted can be divided into the category to be added, and the rest can be divided into the category to be added. Of course, one can also be divided into the category to be added, and the rest can be divided into the category to be added.

[0078] Furthermore, in this embodiment, the above-mentioned device can determine the division ratio according to the quantity information, wherein the division ratio of two time periods can be 1:1, and the division ratio of three time periods can be 1:2 or 2:1.

[0079] Step S23: Divide each of the voice instructions to be broadcast according to the division ratio and the importance, and obtain the category to be added and the category to be added.

[0080] For example, if there are two voice commands to be broadcast, command A and command B, and command A and command B have different importance levels, the above device can classify the one with higher importance as the added voice command, and classify the one with lower importance as the voice command to be added;

[0081] If there are two voice commands to be broadcast, command A and command B, and command A and command B are of the same importance, the above-mentioned device can select the voice commands to be broadcast that contain symbols such as commas and semicolons that separate sentences and classify them into the added category, and classify the ones that do not exist into the category to be added; if both exist or neither exists, the one with fewer words can be transferred to the category to be added, and the one with more words can be classified into the category to be added.

[0082] If there are three voice commands to be broadcasted, command A, command B, and command C, and the importance of the three commands is inconsistent, in accordance with the principle of clarity, the above device can classify the voice commands to be broadcasted with the importance of "important" and "general" into the added category in a 2:1 manner, and classify the voice commands to be broadcasted with the importance of "unimportant" into the to-be-added category;

[0083] If there are three voice commands to be broadcast, command A, command B, and command C, and two of the three commands have the same importance, the division ratio can be selected according to the number of voice commands to be broadcast corresponding to each importance. For example, if two commands are "important" and one is "general" or "unimportant", the ratio can be 2:1. If one command is "important" and the remaining two are "general" or "unimportant", the ratio can be 1:2.

[0084] If there are three voice instructions to be broadcast, namely, instruction A, instruction B and instruction C, and the importance of the three instructions is the same, considering that the number of words in the voice instructions to be broadcast is not large even if it is large, the above-mentioned device in this embodiment can adopt a 2:1 method, that is, the two voice instructions to be broadcast with more words are divided into the added category, and the one with the least words is divided into the category to be added.

[0085] In a specific implementation, the above-mentioned device can determine the importance of each voice instruction to be broadcast according to the semantic analysis result, and determine the division ratio, divide each voice instruction to be broadcast based on the division ratio and importance, and obtain the category to be added and the category to be added.

[0086] Step S30: adding each voice instruction to be added in the to-be-added category to the corresponding voice instruction to be added in the added category.

[0087] It should be noted that the voice instructions to be added may be voice instructions to be broadcasted in the category to be added, and the voice instructions to be added may be voice instructions to be broadcasted in the category to be added.

[0088] After determining the categories corresponding to the voice commands to be broadcast, the voice commands to be added can be added to the added voice commands to achieve command fusion. For example, if "Traffic violation photo ahead, please fasten your seat belt" is the added category command and "One kilometer to destination" is the voice command to be added, then "One kilometer to destination" can be added to "Traffic violation photo ahead, please fasten your seat belt" to obtain the voice command "Traffic violation photo ahead, one kilometer to destination, please fasten your seat belt".

[0089] Step S40: broadcasting each added voice instruction.

[0090] In a specific implementation, the above-mentioned device can broadcast the added voice command after the above-mentioned addition, that is, broadcast "Traffic violation photos are being taken ahead, there is one kilometer left to the destination, please fasten your seat belts". Compared with the existing method of first broadcasting "Traffic violation photos are being taken ahead, please fasten your seat belts" and then broadcasting "One kilometer left to the destination", this embodiment can broadcast "One kilometer left to the destination" in advance, thereby improving the timeliness of the broadcast and improving the user experience.

[0091] Furthermore, since the above device merges the two voice commands to be broadcasted together, in order to facilitate user distinction, the present embodiment can set different broadcast speeds and broadcast emphasis for the unused voice commands to be broadcasted, specifically: before the above step S40, it also includes:

[0092] Step S401: determining a first broadcasting speed and a first broadcasting emphasis of the voice instruction to be added, and determining a second broadcasting speed and a second broadcasting emphasis of the voice instruction to be added;

[0093] Accordingly, the above step S40 includes:

[0094] Step S402: each added voice instruction is broadcasted according to the first broadcasting speed, the second broadcasting speed, the first broadcasting emphasis and the second broadcasting emphasis.

[0095] It should be noted that, in this embodiment, the second broadcast speed can be set lower than the first broadcast speed, and the second broadcast emphasis can be set heavier than the first broadcast emphasis, so that when broadcasting the added part, the above-mentioned device can broadcast at a slower speed and a heavier emphasis, and when broadcasting the part to be added, the above-mentioned device can broadcast at a faster speed and a lighter emphasis.

[0096] For example, when broadcasting the above-mentioned "traffic violation photo ahead", it will be broadcast at the second broadcast speed and the second broadcast emphasis, when broadcasting "one kilometer left to destination", it will be broadcast at the first broadcast speed and the first broadcast emphasis, and when broadcasting "please fasten your seat belts", it will continue to be broadcast at the second broadcast speed and the second broadcast emphasis, so as to facilitate user distinction.

[0097] As a real-time method, the above device can also set different timbres for the voice command to be added and the voice command to be added, and broadcast them with different timbres to facilitate users to distinguish them.

[0098] In this embodiment, when different voice assistants generate voice instructions to be broadcasted at the same time, the above-mentioned device can perform semantic analysis on the received multiple voice instructions to be broadcasted; after performing semantic analysis on each voice instruction to be broadcasted, the above-mentioned device can classify each voice instruction to be broadcasted according to the semantic analysis result, and divide the multiple voice instructions to be broadcasted into a category to be added and a category to be added; after determining the category corresponding to each voice instruction to be broadcasted, the voice instruction to be added can be added to the voice instruction to be added, so as to realize the fusion of the instructions; finally, the added voice instruction is broadcasted. Compared with the existing method of broadcasting according to priority when there are multiple voice instructions, this embodiment can broadcast multiple voice instructions to be broadcasted after adding and fusion, thereby improving the timeliness of the broadcast and improving the user experience.

[0099] refer to Figure 3 , Figure 3 Schematic diagram of the flow chart of the second embodiment of the in-vehicle voice broadcasting method of the present invention.

[0100] Considering that the added voice instruction has a large number of words, the present embodiment can abbreviate the voice instruction to be added for subsequent broadcasting. Figure 3 As shown, in this embodiment, the above step S30 includes:

[0101] Step S31: determining an addable flag of each added voice instruction based on the semantic analysis result of each added voice instruction in the added category.

[0102] It should be noted that the above-mentioned addable flag can be a flag to determine the adding position. In this embodiment, symbols that separate sentences, such as commas and semicolons, can be used as the above-mentioned addable flag. For example, if the added voice instruction is "Traffic violation photo ahead, please fasten your seat belt", the "," between "Traffic violation photo ahead" and "Please fasten your seat belt" can be used as the above-mentioned addable flag.

[0103] Step S32: abbreviating each of the voice instructions to be added in the category to be added according to the semantic analysis result of each of the voice instructions to be added.

[0104] After determining the addable flag in the added voice command, the above device can abbreviate the voice command to be added, wherein the abbreviation can only retain the key information form and remove the redundant words. The specific key information can be set according to the actual situation, and this embodiment does not limit this. For example, the above device can abbreviate "one kilometer left to the destination" and obtain the abbreviated voice command to be added as "one kilometer left".

[0105] As an implementation method, when performing abbreviation, the above-mentioned device can replace the current word with more words with a synonym with fewer words through synonym replacement; for example, "destination" can be replaced with "end point" or "target", and "remaining" can be replaced with "only", etc. This embodiment is not limited to this.

[0106] Step S33: adding each abbreviated voice instruction to be added to the corresponding voice instruction to be added according to each of the addable flags.

[0107] Based on the above example, after adding, you can get the voice command "Traffic violation photo ahead, one kilometer left, please fasten your seat belt."

[0108] Furthermore, considering that there are multiple addable flags in the voice instruction to be added, for example, multiple commas, in order to accurately determine the adding position, in this embodiment, the above step S33 includes:

[0109] Step S331: combining each abbreviated voice instruction to be added with each added voice instruction based on each addable flag to obtain a list of optional combinations.

[0110] It is understandable that the above-mentioned device can determine the combination mode according to each addable flag and generate the above-mentioned optional combination list. Exemplarily, if the added voice instruction is "Speed ​​measurement in the section ahead, speed limit 60, please pay attention to control the speed", then the "," between "Speed ​​measurement in the section ahead" and "Speed ​​limit 60", and the "," between "Speed ​​limit 60" and "Please pay attention to control the speed" can be used as the above-mentioned addable flag. If "The tunnel is about to be entered ahead" is used as the voice instruction to be added, and the abbreviated instruction to be added is "Will enter the tunnel", then the above-mentioned optional combination list may include two combinations, namely: 1. "Speed ​​measurement in the section ahead, will enter the tunnel, speed limit 60, please pay attention to control the speed"; and 2. "Speed ​​measurement in the section ahead, speed limit 60, will enter the tunnel, please pay attention to control the speed".

[0111] Step S332: Determine the text similarity of the added voice instructions after each combination in the optional combination list.

[0112] The above-mentioned text similarity can be the similarity of the context in the combined added voice instructions. Based on the above example, the obtained combined added instructions can be: 1. "Speed ​​measurement in the section ahead, speed limit 60, about to enter the tunnel, please pay attention to control the speed"; and 2. "Speed ​​measurement in the section ahead, speed limit 60, about to enter the tunnel, please pay attention to control the speed"; then the above-mentioned device can respectively calculate the text similarity of the two combined added voice instructions, and analyze the similarity between each paragraph of text.

[0113] Specifically: for 1, the above-mentioned device can analyze the similarities between "speed measurement in the section ahead" and "will enter the tunnel", "will enter the tunnel" and "speed limit 60", "speed limit 60" and "please pay attention to control the speed of the vehicle" in turn; for 2, the above-mentioned device can analyze the similarities between "speed measurement in the section ahead" and "speed limit 60", "speed limit 60" and "will enter the tunnel", "will enter the tunnel" and "please pay attention to control the speed of the vehicle" in turn.

[0114] Step S333: selecting a target combination scheme from the optional combination list based on the text similarities, and adding each abbreviated voice instruction to be added to the corresponding voice instruction to be added based on the target combination scheme.

[0115] It should be understood that after the above-mentioned device obtains the similarity between each segment in turn, it can calculate the sum of the similarities in each solution, and select the solution corresponding to the larger similarity as the above-mentioned target combination solution.

[0116] Based on the above examples, it is not difficult to find that the similarity between "Speed ​​measurement in the section ahead" and "Speed ​​limit 60" is higher than the similarity between "Speed ​​measurement in the section ahead" and "Will enter the tunnel", so Scheme 2 can be used as the above target combination scheme, and "Will enter the tunnel" can be added to "Speed ​​measurement in the section ahead, speed limit 60, please pay attention to control the speed" to obtain the added voice command "Speed ​​measurement in the section ahead, speed limit 60, will enter the tunnel, please pay attention to control the speed".

[0117] In this embodiment, the addable flag in the added voice instruction can be first determined, and the voice instruction to be added can be abbreviated, and the abbreviated voice instruction to be added can be added to the addable flag to obtain the added voice instruction for subsequent broadcasting.

[0118] refer to Figure 4 , Figure 4 Schematic diagram of the flow chart of the third embodiment of the in-vehicle voice broadcasting method of the present invention.

[0119] Considering that the aforementioned embodiments all add the voice command to be added to the voice command to be added, when the difference between the voice command to be added and the voice command to be added is large, context disconnection may occur, which is not easy for users to understand. Figure 4 As shown, in this embodiment, after the above step S33, it also includes:

[0120] Step S34: extracting features from each added voice instruction, and performing context relevance analysis on the extracted features.

[0121] It should be noted that the above device can extract features from the added voice commands and perform context relevance analysis on the extracted features. If we continue to explain based on the above “Traffic violation photo ahead, one kilometer left, please fasten your seat belt”, the above device can extract features from “Traffic violation photo ahead, one kilometer left, please fasten your seat belt”, extract the vocabulary, grammar, and syntax therein, and perform context relevance analysis on the vocabulary, grammar, and syntax.

[0122] The specific context relevance analysis may be performed by cosine similarity calculation or the like, which is not limited in this embodiment. This embodiment uses cosine similarity for illustration, and after the above context relevance analysis, a cosine similarity value may be obtained.

[0123] Step S35: Processing each of the added voice instructions according to the analysis result.

[0124] After determining the cosine similarity value, the above device may remove or supplement the added voice instruction according to the cosine similarity value. Specifically, the above step S35 includes:

[0125] The current association value is determined according to the analysis result, and it is judged whether the current association degree value is lower than a preset association degree threshold; if so, each added voice instruction is supplemented or deleted.

[0126] It should be noted that the current association value is the cosine similarity value between the instruction to be added and the subsequent text in the added voice instruction, that is, the cosine similarity value between "one kilometer remaining" and "fasten your seat belts".

[0127] It should also be noted that the above-mentioned device can be set with a preset correlation degree threshold. If the cosine similarity value is lower than the preset correlation degree threshold, it means that the difference between the two is large, which may cause context disconnection and inconvenience for users to understand. Therefore, the above-mentioned device can supplement or delete "Please fasten your seat belts".

[0128] As for supplementation, in this embodiment, the text before the voice instruction to be added in the added voice instruction can be added after the instruction to be added. The same can be done in abbreviated form. Based on the above example, after supplementation, it can be "Traffic violation photo taking ahead, one kilometer remaining, photo taking ahead, please fasten your seat belts."

[0129] As for deletion, in this embodiment, the text after the to-be-added instruction in the added voice instruction can be deleted. Based on the above example, after supplementary processing, it can be "take a photo of the traffic violation ahead, one kilometer remaining."

[0130] It should be emphasized that the above-mentioned device in this embodiment can determine whether to choose to supplement or delete based on the correlation between the text after the instruction to be added in the added voice instruction and the text before the instruction to be added in the added voice instruction. For example, if the correlation between "Take a photo of the traffic violation ahead" and "Please fasten your seat belts" is relatively large, then the supplement method is selected; if the correlation between "Take a photo of the traffic violation ahead" and "Please fasten your seat belts" is relatively small, then the deletion method is selected.

[0131] In a specific implementation, the above-mentioned device can determine whether to supplement or delete the added voice instruction according to the analysis result of the context relevance degree analysis.

[0132] Accordingly, the above step S40 includes:

[0133] Step S40': announce each processed added voice instruction.

[0134] In a specific implementation, after supplementation or deletion, the above-mentioned device can broadcast the processed added voice instructions.

[0135] In addition, an embodiment of the present invention further proposes a storage medium, on which an in-vehicle voice broadcast program is stored. When the in-vehicle voice broadcast program is executed by a processor, the steps of the in-vehicle voice broadcast method described above are implemented.

[0136] Reference Figure 5 , Figure 5 This is a structural block diagram of the first embodiment of the in-vehicle voice broadcasting device of the present invention.

[0137] like Figure 5 As shown, the in-vehicle voice broadcasting device proposed in the embodiment of the present invention includes:

[0138] The semantic analysis module 501 is used to perform semantic analysis on each of the voice instructions to be broadcast when receiving multiple voice instructions to be broadcast;

[0139] The instruction classification module 502 is used to classify each of the voice instructions to be broadcast according to each semantic analysis result to obtain a category to be added and a category to be added;

[0140] An instruction adding module 503, used for adding each voice instruction to be added in the to-be-added category to the corresponding voice instruction to be added in the added category;

[0141] The voice broadcasting module 504 is used to broadcast each added voice instruction.

[0142] In this embodiment, when different voice assistants generate voice instructions to be broadcasted at the same time, the above-mentioned device can perform semantic analysis on the received multiple voice instructions to be broadcasted; after performing semantic analysis on each voice instruction to be broadcasted, the above-mentioned device can classify each voice instruction to be broadcasted according to the semantic analysis result, and divide the multiple voice instructions to be broadcasted into a category to be added and a category to be added; after determining the category corresponding to each voice instruction to be broadcasted, the voice instruction to be added can be added to the voice instruction to be added, so as to realize the fusion of the instructions; finally, the added voice instruction is broadcasted. Compared with the existing method of broadcasting according to priority when there are multiple voice instructions, this embodiment can broadcast multiple voice instructions to be broadcasted after adding and fusion, thereby improving the timeliness of the broadcast and improving the user experience.

[0143] As an implementation mode, the instruction division module 502 is also used to determine the importance of each voice instruction to be broadcast according to the results of each semantic analysis; obtain the quantity information of each voice instruction to be broadcast, and determine the division ratio based on the quantity information; divide each voice instruction to be broadcast according to the division ratio and the importance, and obtain the category to be added and the category to be added.

[0144] As an implementation manner, the voice broadcast module 504 is further used to determine a first broadcast speed and a first broadcast emphasis of the voice instruction to be added, and determine a second broadcast speed and a second broadcast emphasis of the voice instruction to be added;

[0145] The voice broadcast module 504 is further used to broadcast each added voice instruction according to the first broadcast speed, the second broadcast speed, the first broadcast emphasis and the second broadcast emphasis.

[0146] Based on the first embodiment of the in-vehicle voice broadcasting device of the present invention, a second embodiment of the in-vehicle voice broadcasting device of the present invention is proposed.

[0147] In this embodiment, the instruction adding module 503 is also used to determine the addable flag of each added voice instruction based on the semantic analysis result of each added voice instruction in the added category; abbreviate each voice instruction to be added according to the semantic analysis result of each voice instruction to be added in the to-be-added category; and add each abbreviated voice instruction to be added to the corresponding added voice instruction according to each addable flag.

[0148] As an implementation mode, the instruction adding module 503 is also used to combine each abbreviated voice instruction to be added with each added voice instruction based on each addable flag to obtain an optional combination list; determine the text similarity of each combined added voice instruction in the optional combination list; select a target combination scheme from the optional combination list based on each text similarity, and add each abbreviated voice instruction to be added to the corresponding added voice instruction based on the target combination scheme.

[0149] Based on the above-mentioned embodiments of the in-vehicle voice broadcast device of the present invention, a third embodiment of the in-vehicle voice broadcast device of the present invention is proposed.

[0150] In this embodiment, the instruction adding module 503 is further used to extract features of each added voice instruction, and perform context relevance analysis on the extracted features; and process each added voice instruction according to the analysis results;

[0151] The voice broadcast module 504 is also used to broadcast each processed added voice instruction.

[0152] As an implementation mode, the instruction adding module 503 is also used to determine the current association value according to the analysis result, and judge whether the current association degree value is lower than the preset association degree threshold; if so, each added voice instruction is supplemented or deleted.

[0153] Other embodiments or specific implementations of the in-vehicle voice broadcast device of the present invention can refer to the above-mentioned method embodiments and will not be described in detail here.

[0154] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0155] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0156] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0157] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for in-vehicle voice broadcasting, characterized in that: The method comprises the following steps: When receiving multiple voice instructions to be broadcasted, performing semantic analysis on each of the voice instructions to be broadcasted; Classify the voice instructions to be broadcast according to the semantic analysis results to obtain the category to be added and the category to be added; Adding each voice instruction to be added in the to-be-added category to the corresponding voice instruction to be added in the added category; Each added voice command is announced.

2. The in-vehicle voice broadcast method according to claim 1, characterized in that: The step of adding each voice instruction to be added in the to-be-added category to the corresponding voice instruction to be added in the added category includes: Determine an addable flag of each of the added voice instructions based on a semantic analysis result of each of the added voice instructions in the added category; abbreviating each of the voice instructions to be added according to the semantic analysis result of each of the voice instructions to be added in the category to be added; According to the addable flags, each abbreviated voice instruction to be added is added to the corresponding voice instruction to be added.

3. The in-vehicle voice broadcasting method as claimed in claim 2, characterized in that: The step of adding each abbreviated voice instruction to be added to the corresponding voice instruction to be added according to each of the addable flags includes: Combining each abbreviated voice instruction to be added with each added voice instruction based on each addable flag to obtain a list of optional combinations; Determine the text similarity of the added voice instructions after each combination in the optional combination list; A target combination scheme is selected from the optional combination list based on the text similarities, and each abbreviated voice instruction to be added is added to the corresponding voice instruction to be added based on the target combination scheme.

4. The in-vehicle voice broadcasting method as claimed in claim 2, characterized in that: After the step of adding each abbreviated voice instruction to be added to the corresponding voice instruction to be added according to each addable flag, the method further includes: Extracting features of each added voice command, and analyzing the degree of context relevance of the extracted features; Processing each of the added voice instructions according to the analysis result; Accordingly, the step of broadcasting each added voice instruction includes: The processed added voice commands are announced.

5. The in-vehicle voice broadcasting method as claimed in claim 4, characterized in that: The step of processing each of the added voice instructions according to the analysis result includes: Determine the current correlation value according to the analysis result, and judge whether the current correlation value is lower than a preset correlation threshold; If so, each added voice instruction is supplemented or deleted.

6. The in-vehicle voice broadcast method according to claim 1, characterized in that: The step of dividing the voice instructions to be broadcast according to the semantic analysis results to obtain the categories to be added and the categories to be added includes: Determining the importance of each of the voice instructions to be broadcast according to each semantic analysis result; Obtaining quantity information of each voice command to be broadcast, and determining a division ratio based on the quantity information; The voice instructions to be broadcast are divided according to the division ratio and the importance, and the category to be added and the category to be added are obtained.

7. The in-vehicle voice broadcast method according to any one of claims 1 to 6, characterized in that: Before the step of broadcasting each added voice instruction, the method further includes: Determine a first broadcast speed and a first broadcast emphasis of the voice instruction to be added, and determine a second broadcast speed and a second broadcast emphasis of the voice instruction to be added; Accordingly, the step of broadcasting each added voice instruction includes: Each added voice instruction is announced according to the first announcement speed, the second announcement speed, the first announcement emphasis and the second announcement emphasis.

8. A voice broadcasting device in a car, characterized in that: The device comprises: A semantic analysis module, for performing semantic analysis on each of the voice instructions to be broadcast when receiving multiple voice instructions to be broadcast; An instruction classification module is used to classify each of the voice instructions to be broadcast according to each semantic analysis result to obtain a category to be added and a category to be added; An instruction adding module, used for adding each voice instruction to be added in the to-be-added category to the corresponding voice instruction to be added in the added category; The voice broadcast module is used to broadcast each added voice command.

9. An in-car voice broadcasting device, characterized in that: The device comprises: a memory, a processor, and an in-vehicle voice broadcast program stored in the memory and executable on the processor, wherein the in-vehicle voice broadcast program is configured to implement the steps of the in-vehicle voice broadcast method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores an in-vehicle voice broadcast program, and when the in-vehicle voice broadcast program is executed by the processor, the steps of the in-vehicle voice broadcast method according to any one of claims 1 to 7 are implemented.

Images