Method and device for determining voice robot broadcasting strategy

Abstract

The present invention discloses a method and device for determining the broadcast strategy of a voice robot. The invention comprises: identifying multiple perceptual attributes corresponding to a target conversation process, wherein the multiple perceptual attributes include at least the following attributes: the gender of the target user, the age of the target user, the emotion of the target user, the speech speed of the target user, the appearance characteristics of the target user, the conversation volume of the target user, the background sound of the target conversation process, and the target user corresponds to the target conversation process; determining perceptual attributes of the same level that belong to the same priority, and calculating the final score corresponding to the corresponding priority based on the perceptual attributes of the same level; determining the broadcast strategy of the voice robot based on the final score, wherein the broadcast strategy is at least one of the following: broadcast content, broadcast method. The present invention solves the problem of stiff interaction between the voice robot and the user and poor interaction experience in the related art.

Description

Technical Field

[0001] The present invention relates to the field of robots, and in particular to a method and device for determining a voice robot's broadcasting strategy. Background Art

[0002] In related technologies, the goal of voice robots is to simulate human interaction with users. They must be able to listen (to what the user says), think (to respond to the user), and speak (to the content of their responses). Applications of voice robots include intelligent customer service, smart navigation, smart outbound calls, and voice assistants.

[0003] Existing voice robots primarily utilize three AI technologies: ASR, NLU, and TTS. The conversational interaction process involves: capturing user audio; invoking ASR to convert the audio content into text; invoking NLU to input the recognized text, which is then used to generate a text response based on the NLU knowledge base; invoking TTS to synthesize the response into audio and broadcast it to the user; and repeating this process.

[0004] Existing voice robots use ASR to recognize the user's speech into text. This recognized text is then fed into the NLU, which generates a response and then broadcasts it to the user. The NLU's response depends solely on the text recognized by the ASR. In other words, in the same application scenario, if the knowledge base is the same and the text recognized by the ASR is the same, the subsequent audio broadcast will also be the same. The result is that users who speak the same content will only hear the same response. However, in real-world human interactions, when faced with the same question, we perceive differences in the other person's overall information (gender, age, mood, conversation context, etc.) and provide a more appropriate response. This approach only considers the user's spoken content and fails to account for other factors (such as gender, age, mood, and background noise). This results in a stilted interaction and a poor user experience.

[0005] Currently, no effective solution has been proposed to the above-mentioned problems existing in the related technologies. Summary of the Invention

[0006] The main purpose of the present invention is to provide a method and device for determining the broadcast strategy of a voice robot, so as to solve the problem in the related art that the interaction process between the voice robot and the user is stiff and the interaction experience is poor.

[0007] To achieve the above objectives, according to one aspect of the present invention, a method for determining a speech robot's broadcast strategy is provided. The invention comprises: identifying multiple perceptual attributes corresponding to a target conversation process, wherein the multiple perceptual attributes include at least the following attributes: the target user's gender, the target user's age, the target user's emotion, the target user's speaking speed, the target user's appearance, the target user's conversation volume, and the background sound of the target conversation process, with the target user corresponding to the target conversation process; determining perceptual attributes of the same level that share the same priority level, and calculating a final score corresponding to the corresponding priority level based on the perceptual attributes of the same level; and determining the speech robot's broadcast strategy based on the final score, wherein the broadcast strategy is at least one of the following: broadcast content and broadcast method.

[0008] Furthermore, multiple perceptual attributes corresponding to the target conversation process are identified, including: when the image of the target user cannot be collected, the target voice corresponding to the target conversation process is obtained, and an ASR speech recognizer is called to recognize the target voice to determine multiple perceptual attributes; when the image of the target user can be collected, the target voice corresponding to the target conversation process and the image of the target user are obtained, and an ASR speech recognizer is called to recognize the target voice to determine a first perceptual attribute, and the image of the target user is analyzed to determine a second perceptual attribute, wherein the first perceptual attribute includes at least the following attributes: the target user's emotion, the target user's speaking speed, the target user's conversation volume, and the background sound of the target conversation process, and the second perceptual attribute includes at least the following attributes: the target user's gender, the target user's age, and the target user's appearance characteristics.

[0009] Furthermore, before determining the same-level perceptual attributes belonging to the same priority and calculating the final score corresponding to the corresponding priority based on the same-level perceptual attributes, the method includes: determining the priority, weight and score corresponding to each perceptual attribute.

[0010] Furthermore, the same-level perceptual attributes belonging to the same priority are determined, and the final score corresponding to the corresponding priority is calculated based on the same-level perceptual attributes, including: obtaining a first weight and a first score corresponding to the target user's emotion; obtaining a second weight and a second score corresponding to the background sound of the target conversation process, wherein the target user's emotion and the background sound of the target conversation process belong to the highest priority; calculating a first product between the first weight and the first score; calculating a second product between the second weight and the second score; and determining the ratio between the first sum and the second sum as the final score corresponding to the highest priority, wherein the first sum is the sum of the first product and the second product, and the second sum is the sum between the first score and the second score.

[0011] Furthermore, when the final score is the final score corresponding to the highest priority, the voice robot's broadcast strategy is determined based on the final score, including: judging whether the final score is greater than or equal to a first preset threshold; when the final score corresponding to the highest priority is greater than or equal to the first preset threshold, determining the voice robot's broadcast content to be the first content, and the first content is content used to ask the target user whether it is convenient; when the final score corresponding to the highest priority is less than the first preset threshold, determining the voice robot's broadcast content to be the second content, and the second content is content introducing the business to the target user.

[0012] Furthermore, the same-level perception attributes belonging to the same priority are determined, and the final score corresponding to the corresponding priority is calculated based on the same-level perception attributes, including: obtaining the third weight and third score corresponding to the target user's gender; obtaining the fourth weight and fourth score corresponding to the target user's age; obtaining the fifth weight and fifth score corresponding to the target user's speaking speed; obtaining the sixth weight and sixth score corresponding to the target user's conversation volume, wherein the target user's gender, the target user's age and the customer's conversation volume all belong to secondary priority; calculating the third product between the third weight and the third score; calculating the fourth product between the fourth weight and the fourth score; calculating the fifth product between the fifth weight and the fifth score; calculating the sixth product between the sixth weight and the sixth score; and determining the ratio between the third total and the fourth total as the final score corresponding to the secondary priority, wherein the third total is the sum of the third product, the fourth product, the fifth product and the sixth product, and the fourth total is the sum of the third score, the fourth score, the fifth score and the sixth score.

[0013] Furthermore, based on the final score, the voice robot's broadcasting strategy is determined, including: judging whether the final score is greater than or equal to a second preset threshold; when the final score is greater than or equal to the second preset threshold, determining that the voice robot's broadcasting content is a simplified version of the business introduction; when the final score is less than the second preset threshold, determining that the voice robot's broadcasting content is a detailed version of the business introduction.

[0014] To achieve the above-mentioned objectives, according to another aspect of the present invention, a device for determining a speech robot's broadcast strategy is provided. The device comprises: an identification unit for identifying multiple perceptual attributes corresponding to a target conversation process, wherein the multiple perceptual attributes include at least the following attributes: the target user's gender, the target user's age, the target user's emotion, the target user's speech rate, the target user's appearance, the target user's conversation volume, and the background sound of the target conversation process, and the target user corresponds to the target conversation process; a calculation unit for determining perceptual attributes of the same level that have the same priority level, and calculating a final score corresponding to the corresponding priority level based on the perceptual attributes of the same level; and a first determination unit for determining the speech robot's broadcast strategy based on the final score, wherein the broadcast strategy is at least one of the following: broadcast content and broadcast method.

[0015] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a computer-readable storage medium is provided, which includes a stored program, wherein the program executes any one of the above-mentioned methods for determining a voice robot broadcast strategy.

[0016] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a processor is provided, which is used to run a program, wherein the program executes any one of the above-mentioned methods for determining a voice robot broadcast strategy.

[0017] The present invention employs the following steps: identifying multiple perceptual attributes corresponding to a target conversation process, wherein the multiple perceptual attributes include at least the following attributes: the target user's gender, the target user's age, the target user's emotion, the target user's speech rate, the target user's appearance, the target user's conversation volume, and the background sound of the target conversation process, with the target user corresponding to the target conversation process; determining perceptual attributes of the same level belonging to the same priority, and calculating a final score corresponding to the corresponding priority based on the perceptual attributes of the same level; and determining a voice robot's announcement strategy based on the final score, wherein the announcement strategy includes at least one of the following: announcement content and announcement method. This solves the problem of stiff interaction between voice robots and users and poor interaction experience in related technologies. This results in making the customer experience feel that the interaction process is more like human interaction, thereby improving the customer experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which constitute part of the present invention, are provided to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are provided to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0019] Figure 1 This is a flowchart of a method for determining a voice robot broadcast strategy provided by an embodiment of the present invention;

[0020] Figure 2 is a schematic diagram of a method for determining a voice robot broadcast strategy according to an embodiment of the present invention;

[0021] Figure 3 2 is a schematic diagram of a device for determining a voice robot broadcasting strategy provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0023] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "first," "second," and the like in the specification and claims of the present invention and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate for the embodiments of the present invention described herein. In addition, the terms "including," "having," and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatuses.

[0025] According to an embodiment of the present invention, a method for determining a voice robot announcement strategy is provided.

[0026] Figure 1 Flowchart of a method for determining a voice robot broadcast strategy according to an embodiment of the present invention. Figure 1 As shown, the invention includes the following steps:

[0027] Step S101, identifying multiple perceptual attributes corresponding to the target conversation process, wherein the multiple perceptual attributes include at least the following attributes: gender of the target user, age of the target user, emotion of the target user, speaking speed of the target user, appearance characteristics of the target user, conversation volume of the target user, background sound of the target conversation process, and the target user corresponds to the target conversation process.

[0028] As described above, the present application provides a method for perceiving comprehensive information of a user by acquiring a conversation, wherein the call process can be a call transmitted through a mobile terminal, or it can be a direct conversation process between a voice robot and a user, that is, the application scenario can be the two ends of a wireless call, and at the same time, it can be a direct face-to-face user and a voice robot.

[0029] Specifically, an embodiment of the present application provides a voice robot equipped with an ASR speech recognizer. After acquiring the target conversation process, the voice robot calls the ASR speech recognizer to recognize the text content spoken by the target user, as well as the target user's gender, age, emotion, speech speed, conversation volume, and background sound information. The voice robot uses its own camera to call image recognition and face recognition to recognize the target user's gender, age, and appearance features, including skin color, body shape, and hair. By comprehensively considering the perceptual attributes of the target user obtained by the ASR speech recognizer and camera recognition, a complete perception of the target user can be obtained, that is, a user portrait of the user can be obtained, and the subsequent human-computer voice interaction experience can be improved based on the user portrait.

[0030] Step S102: determining the same-level perception attributes that belong to the same priority level, and calculating the final score corresponding to the corresponding priority level based on the same-level perception attributes.

[0031] As described above, according to the actual application scenario, a priority is set for each perception attribute. The priority can be represented by a number, such as 1, 2, 3, etc. Perception attributes with the same priority number belong to the same level of perception attributes with the same priority. The final score can be calculated based on the priority of the perception attribute.

[0032] Step S103: Determine the announcement strategy of the voice robot based on the final score, wherein the announcement strategy is at least one of the following: announcement content and announcement method.

[0033] As described above, the final score is compared with different thresholds, and the content and method of the voice robot's broadcast are determined based on the relationship between the final score and the threshold.

[0034] Specifically, in the embodiments provided in the present application, in an optional scenario provided, if the voice robot is used to provide information packages for users, the broadcast content includes a simplified version and a detailed version of the package, and the broadcast method includes the broadcast voices of boys and girls, etc.

[0035] In an optional instance, multiple perceptual attributes corresponding to a target conversation process are identified, including: when an image of the target user cannot be captured, obtaining a target voice corresponding to the target conversation process, calling an ASR speech recognizer to recognize the target voice to determine multiple perceptual attributes; when an image of the target user can be captured, obtaining a target voice corresponding to the target conversation process and an image of the target user, calling an ASR speech recognizer to recognize the target voice to determine a first perceptual attribute, and analyzing the image of the target user to determine a second perceptual attribute, wherein the first perceptual attribute includes at least the following attributes: the target user's emotion, the target user's speaking speed, the target user's conversation volume, and the background sound of the target conversation process, and the second perceptual attribute includes at least the following attributes: the target user's gender, the target user's age, and the target user's appearance characteristics.

[0036] As mentioned above, this application provides a variety of scenarios for interacting with voice robots. In the scenario of interacting with the voice robot through a communication terminal, it is impossible to obtain the user's image. Therefore, when the image of the target user cannot be collected, the voice robot recognizes the customer's voice through the ASR speech recognizer. The ASR speech recognizer can not only recognize the text content, but also recognize the user's gender, age, emotion, speaking speed, background sound and other attributes.

[0037] When used for direct communication with a voice robot, the voice robot is equipped with a camera that can capture images of the user. Therefore, when capturing the image of the target user, it is necessary not only to call the ASR speech recognizer to recognize the target voice, but also to capture the customer's image through the camera and use the voice robot's built-in camera to identify the user's gender, age, height, body shape, skin color, hair and other attributes.

[0038] In an optional example, before determining the same-level perceptual attributes belonging to the same priority and calculating the final score corresponding to the corresponding priority based on the same-level perceptual attributes, the method includes: determining the priority, weight and score corresponding to each perceptual attribute.

[0039] As described above, priorities, weights, and scores are set for each perception attribute based on the actual application scenario.

[0040] Table 1 is an optional embodiment provided according to the present invention, which provides a setting table of the priority, weight, and score of the perception attributes of a virtual robot involved in a certain financial business. The smaller the numerical value of the priority in Table 1, the higher the priority, and a weight of 0 means that this perception attribute value is not considered. Table 1 is shown in the following table.

[0041]

[0042]

[0043] Table 2 is another embodiment provided according to the present invention, which provides another table of perception information and score settings, as shown in Figure 2. Table 2 is shown in the following table:

[0044]

[0045]

[0046] In an optional instance, the same-level perceptual attributes belonging to the same priority are determined, and the final score corresponding to the corresponding priority is calculated based on the same-level perceptual attributes, including: obtaining a first weight and a first score corresponding to the target user's emotion; obtaining a second weight and a second score corresponding to the background sound of the target conversation process, wherein the target user's emotion and the background sound of the target conversation process belong to the highest priority; calculating a first product between the first weight and the first score; calculating a second product between the second weight and the second score; and determining the ratio between the first sum and the second sum as the final score corresponding to the highest priority, wherein the first sum is the sum of the first product and the second product, and the second sum is the sum between the first score and the second score.

[0047] In an optional instance, when the final score is the final score corresponding to the highest priority, the voice robot's broadcast strategy is determined based on the final score, including: judging whether the final score is greater than or equal to a first preset threshold; when the final score corresponding to the highest priority is greater than or equal to the first preset threshold, determining the voice robot's broadcast content to be the first content, and the first content is content used to ask the target user whether it is convenient; when the final score corresponding to the highest priority is less than the first preset threshold, determining the voice robot's broadcast content to be the second content, and the second content is content introducing the business to the target user.

[0048] Specifically, the final score calculation formula for the perceptual attributes of the same priority is:

[0049] PTotal=P1+P2+...+Pn Formula 1

[0050] Final=(S1*P1+S2*P2+...+Si*Pn) / PTotal Formula 2

[0051] Among them, Final: final score, pi: weight of perception attribute i, Si: score of perception attribute i, PTotal: total weight value.

[0052] Step 1: Taking the values ​​in Table 1 as an example, calculate the perception attribute score with priority 1 in the table:

[0053] Final_max = (S background sound_max * P background sound + S emotion_max * P emotion)

[0054] / (P_background sound + P_emotion) = (4*2+3*3) / (3+2) = 3.4

[0055] Final_min = (S background sound_min * P background sound + S emotion_min * P emotion)

[0056] / (P_background sound + P_emotion) = (1*2+1*3) / (3+2) = 1

[0057] Final = (S background sound_current background sound*P background sound+S emotion_current emotion*P emotion)

[0058] / (P_background sound + P_emotion) = (3*2+2*3) / (3+2) = 2.4

[0059] It should be noted that, by calculating the maximum score and the minimum score of the perceptual attribute with a priority of 1, the size of the preset threshold to be compared with the final score can be determined.

[0060] Step 2: Taking the values ​​in Table 1 as an example, calculate the perception attribute scores with priority 2 in the table:

[0061] Final_max=(2*1+4*3+3*1+3*1) / (1+3+1+1)=3.33

[0062] Final_min=(1*1+1*3+1*1+1*1) / (1+3+1+1)=1

[0063] Final=(2*1+3*3+2*1+2*1) / (1+3+1+1)=2.5

[0064] Taking the values ​​in Table 1 as an example, the score range for the perceptual attribute with priority 1 in Table 1 is [1, 3.4]. Preferably, the first preset threshold is 2.2. The first preset threshold can be set according to different scenarios. If the final score is greater than the first preset threshold, the voice robot determines that the customer may not be able to answer the phone at this time and will announce to the customer, "Is it convenient for you to answer the phone now?" If the customer responds, "No," the robot will announce, "We'll contact you later. Goodbye," and end the announcement. If the customer responds, "Yes," the robot will continue with the service announcement. If the final score is less than the first preset threshold, the robot will continue with the service announcement.

[0065] In an optional instance, the same-level perception attributes belonging to the same priority are determined, and the final score corresponding to the corresponding priority is calculated based on the same-level perception attributes, including: obtaining a third weight and a third score corresponding to the target user's gender; obtaining a fourth weight and a fourth score corresponding to the target user's age; obtaining a fifth weight and a fifth score corresponding to the target user's speaking speed; obtaining a sixth weight and a sixth score corresponding to the target user's conversation volume, wherein the target user's gender, the target user's age, and the customer's conversation volume all belong to secondary priorities; calculating a third product between the third weight and the third score; calculating a fourth product between the fourth weight and the fourth score; calculating a fifth product between the fifth weight and the fifth score; calculating a sixth product between the sixth weight and the sixth score; and determining the ratio between the third sum and the fourth sum as the final score corresponding to the secondary priority, wherein the third sum is the sum of the third product, the fourth product, the fifth product, and the sixth product, and the fourth sum is the sum of the third score, the fourth score, the fifth score, and the sixth score.

[0066] In an optional instance, the voice robot's broadcast strategy is determined based on the final score, including: judging whether the final score is greater than or equal to a second preset threshold; when the final score is greater than or equal to the second preset threshold, determining that the voice robot's broadcast content is a simplified version of the business introduction; when the final score is less than the second preset threshold, determining that the voice robot's broadcast content is a detailed version of the business introduction.

[0067] As mentioned above, Table 1 is a setting table of the priority, weight, and score of the perception attributes of a virtual robot involved in a certain financial business provided according to an embodiment of the present invention. Based on the data in Table 1, Formula 1 and Formula 2, the weights and final scores corresponding to the perception attributes of the target user can be calculated respectively.

[0068] As mentioned above, taking the values ​​in Table 1 as an example, the score range of the perception attribute with priority 2 in Table 1 is [1, 3.33]. Preferably, the second preset threshold is 1.2. The second preset threshold can be set according to different scenarios. If the final score is greater than or equal to the second preset threshold, a simplified version of the business introduction is broadcast to the customer. If the final score is less than the second preset threshold, a detailed version of the business introduction is broadcast to the customer.

[0069] In an alternative embodiment of the present invention, for example, an outbound call robot recommends mobile phone plans to a customer. Based on its understanding of the customer's information during voice interaction, the robot recommends a plan that best suits the customer. For example, it recommends plans with high data usage to young people, while recommends plans with low data usage but low overall costs to seniors.

[0070] If the background sound contains car horns, and you judge that the customer may be driving, you can first ask the customer whether he is driving and remind him to drive safely.

[0071] If background noise is loud or multiple voices are detected, you can first ask the customer if they are available to answer the call. If so, play a simplified version of the package introduction. If the customer says no, skip the package introduction and inform them that you will contact them later. If the customer is impatient, play a simplified version of the package introduction. If the customer is calm and speaking at a normal speed, play a detailed version of the package introduction.

[0072] Taking the physical robot placed in the shopping mall as an example, the robot can recommend suitable cosmetics to the customer based on the perceived gender, age, and skin color, or inform them of the floor information and promotional information suitable for their shopping.

[0073] An embodiment of the present invention provides a method for determining a voice robot's announcement strategy. The method identifies multiple perceptual attributes corresponding to a target conversation process, wherein the multiple perceptual attributes include at least the following: the target user's gender, age, emotion, speech rate, appearance, volume, and background sound of the target conversation process; the target user corresponds to the target conversation process; determines perceptual attributes of the same level with the same priority, and calculates a final score corresponding to the corresponding priority based on the perceptual attributes of the same level; and determines the voice robot's announcement strategy based on the final score, wherein the announcement strategy includes at least one of the following: announcement content and announcement method. This method solves the problem of stiff interaction between voice robots and users and poor interaction experience in related technologies. This method makes the customer experience feel that the interaction process is more similar to human interaction, thereby improving the customer experience.

[0074] The above-mentioned places, such as Figure 2 As shown, Figure 2 This is a schematic diagram of a method for determining a voice robot's broadcast strategy provided according to an embodiment of the present invention. The voice robot can automatically perceive the comprehensive information of the target user. When deciding the response content, it will consider what the target user said and the comprehensive information of the target user himself, and give a more appropriate answer content. The multi-person and multi-faceted approach makes the user feel that the interaction process is closer to human interaction, no longer mechanical and stiff, and gives the user a better interactive experience.

[0075] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0076] An embodiment of the present invention further provides a device for determining a voice robot's broadcast strategy. It should be noted that the device for determining a voice robot's broadcast strategy according to an embodiment of the present invention can be used to execute the method for determining a voice robot's broadcast strategy provided in an embodiment of the present invention. The following describes the device for determining a voice robot's broadcast strategy according to an embodiment of the present invention.

[0077] Figure 3 Schematic diagram of a device for determining a voice robot broadcast strategy according to an embodiment of the present invention. Figure 3 As shown, the device includes: an identification unit 301, which is used to identify multiple perceptual attributes corresponding to the target dialogue process, wherein the multiple perceptual attributes include at least the following attributes: the gender of the target user, the age of the target user, the emotion of the target user, the speaking speed of the target user, the appearance characteristics of the target user, the conversation volume of the target user, the background sound of the target dialogue process, and the target user corresponds to the target dialogue process; a calculation unit 302, which is used to determine the same-level perceptual attributes belonging to the same priority, and calculate the final score corresponding to the corresponding priority based on the same-level perceptual attributes; a first determination unit 303, which is used to determine the broadcast strategy of the voice robot based on the final score, wherein the broadcast strategy is at least one of the following: broadcast content and broadcast method.

[0078] In an optional instance, the recognition unit 301 includes: a first recognition subunit, which is used to obtain the target voice corresponding to the target conversation process when the image of the target user cannot be collected, and call the ASR speech recognizer to recognize the target voice to determine multiple perceptual attributes; a second recognition subunit, which is used to obtain the target voice corresponding to the target conversation process and the image of the target user when the image of the target user can be collected, call the ASR speech recognizer to recognize the target voice to determine the first perceptual attribute, and analyze the image of the target user to determine the second perceptual attribute, wherein the first perceptual attribute includes at least the following attributes: the target user's emotion, the target user's speaking speed, the target user's conversation volume, and the background sound of the target conversation process, and the second perceptual attribute includes at least the following attributes: the target user's gender, the target user's age, and the target user's appearance characteristics.

[0079] In an optional instance, the device includes: a second determination unit, which is used to determine the priority, weight and score corresponding to each perception attribute before determining the same-level perception attributes belonging to the same priority and calculating the final score corresponding to the corresponding priority based on the same-level perception attributes.

[0080] In an optional instance, the calculation unit 302 includes: a first acquisition subunit, used to obtain a first weight and a first score corresponding to the target user's emotion; a second acquisition subunit, used to obtain a second weight and a second score corresponding to the background sound of the target conversation process, wherein the target user's emotion and the background sound of the target conversation process both belong to the highest priority; a first calculation subunit, used to calculate a first product between the first weight and the first score; a second calculation subunit, used to calculate a second product between the second weight and the second score; a first determination subunit, used to determine the ratio between the first sum and the second sum as the final score corresponding to the highest priority, wherein the first sum is the sum of the first product and the second product, and the second sum is the sum between the first score and the second score.

[0081] In an optional instance, the first determination unit 303 includes: a first judgment subunit, used to determine whether the final score is greater than or equal to a first preset threshold; a second determination subunit, used to determine that the broadcast content of the voice robot is the first content when the final score corresponding to the highest priority is greater than or equal to the first preset threshold, and the first content is content for asking the target user whether it is convenient; a third determination subunit, used to determine that the broadcast content of the voice robot is the second content when the final score corresponding to the highest priority is less than the first preset threshold, and the second content is content introducing the business to the target user.

[0082] In an optional instance, the calculation unit 302 includes: a third acquisition subunit, used to obtain a third weight and a third score corresponding to the gender of the target user; a fourth acquisition subunit, used to obtain a fourth weight and a fourth score corresponding to the age of the target user; a fifth acquisition subunit, used to obtain a fifth weight and a fifth score corresponding to the speaking speed of the target user; a sixth acquisition subunit, used to obtain a sixth weight and a sixth score corresponding to the conversation volume of the target user, wherein the gender of the target user, the age of the target user, and the conversation volume of the customer all belong to secondary priorities; a third calculation subunit, used to calculate a third product between the third weight and the third score; a fourth calculation subunit, used to calculate a fourth product between the fourth weight and the fourth score; a fifth calculation subunit, used to calculate a fifth product between the fifth weight and the fifth score; a sixth calculation subunit, used to calculate a sixth product between the sixth weight and the sixth score; a fourth determination subunit, used to determine the ratio between the third sum and the fourth sum as the final score corresponding to the secondary priority, wherein the third sum is the sum of the third product, the fourth product, the fifth product, and the sixth product, and the fourth sum is the sum of the third score, the fourth score, the fifth score, and the sixth score.

[0083] In an optional instance, the first determination unit 303 includes: a second judgment subunit, used to determine whether the final score is greater than or equal to a second preset threshold; a fifth determination subunit, used to determine that the voice robot's broadcast content is a simplified version of the business introduction when the final score is greater than or equal to the second preset threshold; and a sixth determination subunit, used to determine that the voice robot's broadcast content is a detailed version of the business introduction when the final score is less than the second preset threshold.

[0084] An embodiment of the present invention provides a device for determining a voice robot's broadcast strategy, which identifies multiple perceptual attributes corresponding to a target dialogue process, wherein the multiple perceptual attributes include at least the following attributes: the gender of a target user, the age of a target user, the emotion of a target user, the speaking speed of a target user, the appearance characteristics of a target user, the conversation volume of a target user, and the background sound of a target dialogue process, and the target user corresponds to a target dialogue process; determines perceptual attributes of the same level that belong to the same priority, and calculates a final score corresponding to the corresponding priority based on the perceptual attributes of the same level; determines the voice robot's broadcast strategy based on the final score, wherein the broadcast strategy is at least one of the following: broadcast content and broadcast method, thereby solving the problems of a stiff interaction process and poor interaction experience between the voice robot and the user in related technologies, thereby making the customer feel that the interaction process is closer to interaction with people, thereby improving the customer experience effect.

[0085] The device for determining the voice robot broadcast strategy includes a processor and a memory. The above-mentioned recognition unit 301 and the like are stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0086] The processor contains a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be configured, and kernel parameters can be adjusted to address the issues in related technologies such as the awkward interaction process between voice robots and users, resulting in a poor user experience.

[0087] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0088] An embodiment of the present invention provides a storage medium having a program stored thereon, which, when executed by a processor, implements the method for determining a voice robot broadcast strategy.

[0089] An embodiment of the present invention provides a processor, which is used to run a program, wherein the method for determining a voice robot broadcast strategy is executed when the program is running.

[0090] An embodiment of the present invention provides a device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the following steps are implemented: identifying multiple perceptual attributes corresponding to a target conversation process, wherein the multiple perceptual attributes include at least the following attributes: the gender of a target user, the age of a target user, the emotion of a target user, the speaking speed of a target user, the appearance characteristics of a target user, the conversation volume of a target user, the background sound of a target conversation process, and the target user corresponds to a target conversation process; determining perceptual attributes of the same level that belong to the same priority, and calculating a final score corresponding to the corresponding priority based on the perceptual attributes of the same level; and determining a broadcasting strategy of a voice robot based on the final score, wherein the broadcasting strategy is at least one of the following: broadcast content and broadcasting method.

[0091] In an optional instance, multiple perceptual attributes corresponding to a target conversation process are identified, including: when an image of the target user cannot be captured, obtaining a target voice corresponding to the target conversation process, calling an ASR speech recognizer to recognize the target voice to determine multiple perceptual attributes; when an image of the target user can be captured, obtaining a target voice corresponding to the target conversation process and an image of the target user, calling an ASR speech recognizer to recognize the target voice to determine a first perceptual attribute, and analyzing the image of the target user to determine a second perceptual attribute, wherein the first perceptual attribute includes at least the following attributes: the target user's emotion, the target user's speaking speed, the target user's conversation volume, and the background sound of the target conversation process, and the second perceptual attribute includes at least the following attributes: the target user's gender, the target user's age, and the target user's appearance characteristics.

[0092] In an optional example, before determining the same-level perceptual attributes belonging to the same priority and calculating the final score corresponding to the corresponding priority based on the same-level perceptual attributes, the method includes: determining the priority, weight and score corresponding to each perceptual attribute.

[0093] In an optional instance, the same-level perceptual attributes belonging to the same priority are determined, and the final score corresponding to the corresponding priority is calculated based on the same-level perceptual attributes, including: obtaining a first weight and a first score corresponding to the target user's emotion; obtaining a second weight and a second score corresponding to the background sound of the target conversation process, wherein the target user's emotion and the background sound of the target conversation process belong to the highest priority; calculating a first product between the first weight and the first score; calculating a second product between the second weight and the second score; and determining the ratio between the first sum and the second sum as the final score corresponding to the highest priority, wherein the first sum is the sum of the first product and the second product, and the second sum is the sum between the first score and the second score.

[0094] In an optional instance, when the final score is the final score corresponding to the highest priority, the voice robot's broadcast strategy is determined based on the final score, including: judging whether the final score is greater than or equal to a first preset threshold; when the final score corresponding to the highest priority is greater than or equal to the first preset threshold, determining the voice robot's broadcast content to be the first content, and the first content is content used to ask the target user whether it is convenient; when the final score corresponding to the highest priority is less than the first preset threshold, determining the voice robot's broadcast content to be the second content, and the second content is content introducing the business to the target user.

[0095] In an optional instance, the same-level perception attributes belonging to the same priority are determined, and the final score corresponding to the corresponding priority is calculated based on the same-level perception attributes, including: obtaining a third weight and a third score corresponding to the target user's gender; obtaining a fourth weight and a fourth score corresponding to the target user's age; obtaining a fifth weight and a fifth score corresponding to the target user's speaking speed; obtaining a sixth weight and a sixth score corresponding to the target user's conversation volume, wherein the target user's gender, the target user's age, and the customer's conversation volume all belong to secondary priorities; calculating a third product between the third weight and the third score; calculating a fourth product between the fourth weight and the fourth score; calculating a fifth product between the fifth weight and the fifth score; calculating a sixth product between the sixth weight and the sixth score; and determining the ratio between the third sum and the fourth sum as the final score corresponding to the secondary priority, wherein the third sum is the sum of the third product, the fourth product, the fifth product, and the sixth product, and the fourth sum is the sum of the third score, the fourth score, the fifth score, and the sixth score.

[0096] In an optional instance, the voice robot's broadcast strategy is determined based on the final score, including: judging whether the final score is greater than or equal to a second preset threshold; when the final score is greater than or equal to the second preset threshold, determining that the voice robot's broadcast content is a simplified version of the business introduction; when the final score is less than the second preset threshold, determining that the voice robot's broadcast content is a detailed version of the business introduction.

[0097] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0098] The present invention also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program initialized with the following method steps: identifying multiple perceptual attributes corresponding to a target dialogue process, wherein the multiple perceptual attributes include at least the following attributes: the gender of the target user, the age of the target user, the emotion of the target user, the speaking speed of the target user, the appearance characteristics of the target user, the conversation volume of the target user, the background sound of the target dialogue process, and the target user corresponds to the target dialogue process; determining the same-level perceptual attributes belonging to the same priority, and calculating the final score corresponding to the corresponding priority based on the same-level perceptual attributes; determining the broadcast strategy of the voice robot based on the final score, wherein the broadcast strategy is at least one of the following: broadcast content and broadcast method.

[0099] In an optional instance, multiple perceptual attributes corresponding to a target conversation process are identified, including: when an image of the target user cannot be captured, obtaining a target voice corresponding to the target conversation process, calling an ASR speech recognizer to recognize the target voice to determine multiple perceptual attributes; when an image of the target user can be captured, obtaining a target voice corresponding to the target conversation process and an image of the target user, calling an ASR speech recognizer to recognize the target voice to determine a first perceptual attribute, and analyzing the image of the target user to determine a second perceptual attribute, wherein the first perceptual attribute includes at least the following attributes: the target user's emotion, the target user's speaking speed, the target user's conversation volume, and the background sound of the target conversation process, and the second perceptual attribute includes at least the following attributes: the target user's gender, the target user's age, and the target user's appearance characteristics.

[0100] In an optional example, before determining the same-level perceptual attributes belonging to the same priority and calculating the final score corresponding to the corresponding priority based on the same-level perceptual attributes, the method includes: determining the priority, weight and score corresponding to each perceptual attribute.

[0101] In an optional instance, the same-level perceptual attributes belonging to the same priority are determined, and the final score corresponding to the corresponding priority is calculated based on the same-level perceptual attributes, including: obtaining a first weight and a first score corresponding to the target user's emotion; obtaining a second weight and a second score corresponding to the background sound of the target conversation process, wherein the target user's emotion and the background sound of the target conversation process belong to the highest priority; calculating a first product between the first weight and the first score; calculating a second product between the second weight and the second score; and determining the ratio between the first sum and the second sum as the final score corresponding to the highest priority, wherein the first sum is the sum of the first product and the second product, and the second sum is the sum between the first score and the second score.

[0102] In an optional instance, when the final score is the final score corresponding to the highest priority, the voice robot's broadcast strategy is determined based on the final score, including: judging whether the final score is greater than or equal to a first preset threshold; when the final score corresponding to the highest priority is greater than or equal to the first preset threshold, determining the voice robot's broadcast content to be the first content, and the first content is content used to ask the target user whether it is convenient; when the final score corresponding to the highest priority is less than the first preset threshold, determining the voice robot's broadcast content to be the second content, and the second content is content introducing the business to the target user.

[0103] In an optional instance, the same-level perception attributes belonging to the same priority are determined, and the final score corresponding to the corresponding priority is calculated based on the same-level perception attributes, including: obtaining a third weight and a third score corresponding to the target user's gender; obtaining a fourth weight and a fourth score corresponding to the target user's age; obtaining a fifth weight and a fifth score corresponding to the target user's speaking speed; obtaining a sixth weight and a sixth score corresponding to the target user's conversation volume, wherein the target user's gender, the target user's age, and the customer's conversation volume all belong to secondary priorities; calculating a third product between the third weight and the third score; calculating a fourth product between the fourth weight and the fourth score; calculating a fifth product between the fifth weight and the fifth score; calculating a sixth product between the sixth weight and the sixth score; and determining the ratio between the third sum and the fourth sum as the final score corresponding to the secondary priority, wherein the third sum is the sum of the third product, the fourth product, the fifth product, and the sixth product, and the fourth sum is the sum of the third score, the fourth score, the fifth score, and the sixth score.

[0104] In an optional instance, the voice robot's broadcast strategy is determined based on the final score, including: judging whether the final score is greater than or equal to a second preset threshold; when the final score is greater than or equal to the second preset threshold, determining that the voice robot's broadcast content is a simplified version of the business introduction; when the final score is less than the second preset threshold, determining that the voice robot's broadcast content is a detailed version of the business introduction.

[0105] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0106] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0107] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0108] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0109] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0110] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0111] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0112] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0113] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0114] The above are merely embodiments of the present invention and are not intended to limit the present invention. It will be apparent to those skilled in the art that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims of the present invention.

Claims

1. A method for determining a voice robot's broadcast strategy, characterized in that: include: Identify multiple perceptual attributes corresponding to the target conversation process; Determining perceptual attributes of the same level that belong to the same priority, and calculating a final score corresponding to the corresponding priority based on the perceptual attributes of the same level, wherein the priority is divided into at least a highest priority and a secondary priority, wherein the perceptual attributes corresponding to the highest priority are the target user's emotion and the background sound of the target conversation process, and the perceptual attributes corresponding to the secondary priority are the target user's gender, the target user's age, the target user's conversation volume, and the target user's speaking speed; Determine whether the final score corresponding to the highest priority is greater than or equal to a first preset threshold; if the final score corresponding to the highest priority is greater than or equal to the first preset threshold, determine that the broadcast content of the voice robot is the first content, and the first content is content for asking the target user whether it is convenient; if the final score corresponding to the highest priority is less than the first preset threshold, determine that the broadcast content of the voice robot is the second content, and the second content is content for introducing the service to the target user. At the same time, determine whether the final score corresponding to the secondary priority is greater than or equal to the second preset threshold; if the final score corresponding to the secondary priority is greater than or equal to the second preset threshold, determine that the broadcast content of the voice robot is a simplified version of the service introduction; if the final score corresponding to the secondary priority is less than the second preset threshold, determine that the broadcast content of the voice robot is a detailed version of the service introduction. The first preset threshold is determined according to the score range of the perceptual attribute corresponding to the highest priority, and the second preset threshold is determined according to the score range of the perceptual attribute corresponding to the secondary priority.

2. The method according to claim 1, characterized in that Identify multiple perceptual attributes corresponding to the target conversation process, including: In the case where the image of the target user cannot be collected, obtaining a target speech corresponding to the target conversation process, calling an ASR speech recognizer to recognize the target speech to determine the plurality of perceptual attributes; When the image of the target user can be collected, the target voice corresponding to the target conversation process and the image of the target user are obtained, an ASR speech recognizer is called to recognize the target voice to determine a first perceptual attribute, and the image of the target user is analyzed to determine a second perceptual attribute, wherein the first perceptual attribute includes at least the following attributes: the target user's emotion, the target user's speaking speed, the target user's conversation volume, and the background sound of the target conversation process; and the second perceptual attribute includes at least the following attributes: the target user's gender, the target user's age, and the target user's appearance characteristics.

3. The method according to claim 1, characterized in that Before determining the same-level perception attributes belonging to the same priority and calculating the final score corresponding to the corresponding priority based on the same-level perception attributes, the method includes: Determine the priority, weight, and score corresponding to each of the perception attributes.

4. The method according to claim 2, characterized in that Determining peer perception attributes belonging to the same priority level, and calculating a final score corresponding to the corresponding priority level based on the peer perception attributes, including: Obtaining a first weight and a first score corresponding to the emotion of the target user; Obtaining a second weight and a second score corresponding to the background sound of the target conversation process, wherein the emotion of the target user and the background sound of the target conversation process both have the highest priority; calculating a first product between the first weight and the first score; calculating a second product between the second weight and the second score; The ratio between the first sum and the second sum is determined as the final score corresponding to the highest priority, wherein the first sum is the sum of the first product and the second product, and the second sum is the sum of the first weight and the second weight.

5. The method according to claim 1, wherein Determining peer perception attributes belonging to the same priority level, and calculating a final score corresponding to the corresponding priority level based on the peer perception attributes, including: Obtain a third weight and a third score corresponding to the gender of the target user; Obtaining a fourth weight and a fourth score corresponding to the age of the target user; Obtaining a fifth weight and a fifth score corresponding to the target user's speaking speed; Obtaining a sixth weight and a sixth score corresponding to the target user's conversation volume, wherein the target user's gender, the target user's age, and the target user's conversation volume all belong to secondary priorities; calculating a third product between the third weight and the third score; calculating a fourth product between the fourth weight and the fourth score; calculating a fifth product between the fifth weight and the fifth score; calculating a sixth product between the sixth weight and the sixth score; The ratio between the third sum and the fourth sum is determined as the final score corresponding to the secondary priority, wherein the third sum is the sum of the third product, the fourth product, the fifth product and the sixth product, and the fourth sum is the sum of the third weight, the fourth weight, the fifth weight and the sixth weight.

6. A device for determining a voice robot's broadcast strategy, characterized in that: include: A recognition unit, used to identify multiple perceptual attributes corresponding to the target dialogue process; a calculation unit, configured to determine perceptual attributes of the same level belonging to the same priority, and calculate a final score corresponding to the corresponding priority based on the perceptual attributes of the same level, wherein the priority is divided into at least a highest priority and a secondary priority, wherein the perceptual attributes corresponding to the highest priority are the target user's emotion and the background sound of the target conversation process, and the perceptual attributes corresponding to the secondary priority are the target user's gender, the target user's age, the target user's conversation volume, and the target user's speaking speed; The first determination unit is used to determine whether the final score corresponding to the highest priority is greater than or equal to a first preset threshold; if the final score corresponding to the highest priority is greater than or equal to the first preset threshold, determine that the broadcast content of the voice robot is the first content, and the first content is content for asking the target user whether it is convenient; if the final score corresponding to the highest priority is less than the first preset threshold, determine that the broadcast content of the voice robot is the second content, and the second content is content introducing the service to the target user. At the same time, determine whether the final score corresponding to the secondary priority is greater than or equal to the second preset threshold; if the final score corresponding to the secondary priority is greater than or equal to the second preset threshold, determine that the broadcast content of the voice robot is a simplified version of the service introduction; if the final score corresponding to the secondary priority is less than the second preset threshold, determine that the broadcast content of the voice robot is a detailed version of the service introduction. The first preset threshold is determined according to the score range of the perceptual attribute corresponding to the highest priority, and the second preset threshold is determined according to the score range of the perceptual attribute corresponding to the secondary priority.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute a method for determining a voice robot broadcast strategy as described in any one of claims 1 to 5.

8. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes a method for determining a voice robot broadcast strategy as described in any one of claims 1 to 5.

Images