Information processing method and device and computer equipment
By acquiring descriptive information and contextual data from both parties in the dialogue, and instructing an AI model to simulate role identities and generate target dialogues, the problem of inaccurate dialogue communication in existing technologies is solved, and dialogue generation that better meets user needs is achieved.
Patent Information
- Application Number
- CN202511491576.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-19
- Publication Date
- 2026-02-24
AI Technical Summary
In existing technologies, the responses generated by AI-assisted dialogue methods are not accurate enough and cannot meet the diverse needs of users.
By acquiring descriptive information and contextual data from both parties in the dialogue, the AI model is instructed to simulate role identities, generate target dialogues, and consider factors such as intimacy, tone, and relationship to improve the accuracy of the dialogues.
The generated target dialogue is more in line with user needs and can more accurately express user intent in different scenarios.
Smart Images
Figure CN121561027A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present invention relate to the field of artificial intelligence, and more particularly to an information processing method, an information processing apparatus, and a computer device. Background Technology
[0002] Human society needs to communicate in various languages to achieve daily emotional exchange, work, and other purposes.
[0003] To improve communication efficiency, with the development of technologies such as artificial intelligence and machine learning algorithms, some have attempted to use AI to assist users in conversation. Current assistance methods simply provide a single sentence of dialogue, and the algorithm generates a single response. Due to a lack of further dialogue information, the responses generated by these algorithms are often inaccurate and fail to meet the needs of a wider range of users. Summary of the Invention
[0004] This invention provides an information processing method, an information processing apparatus, and a computer device for generating dialogues that better meet user needs.
[0005] To achieve this objective, the embodiments of the present invention adopt the following technical solutions:
[0006] An information processing method, comprising:
[0007] Obtain the description information of the first character and the description information of the second character;
[0008] Obtain dialogue context data;
[0009] The artificial intelligence model is instructed to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, and the dialogue context data.
[0010] Optionally, the information processing method further includes:
[0011] Obtain the target intimacy level, which is used to represent the degree of intimacy;
[0012] The instruction artificial intelligence model, referring to the description information of the first role, the description information of the second role, and the dialogue context data, simulates the identity of the first role and generates a target dialogue to be spoken to the second role, including:
[0013] The artificial intelligence model is instructed to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, and the dialogue context data.
[0014] The target dialogue corresponds to the level of intimacy described by the target intimacy.
[0015] Optionally, the information processing method further includes:
[0016] Obtain tone type information, which is used to describe the tone type of the speech;
[0017] The instruction artificial intelligence model, referring to the description information of the first character, the description information of the second character, the target intimacy level, and the dialogue context data, simulates the identity of the first character and generates a target dialogue to be spoken to the second character, including:
[0018] The artificial intelligence model is instructed to simulate the identity of the first character by referring to the description information of the first character, the description information of the second character, the target intimacy, the tone type information, and the dialogue context data, and to generate a target dialogue to be spoken to the second character using the tone type described by the tone type information.
[0019] Optionally, the information processing method further includes:
[0020] Obtain relationship information, which describes the relationship between the first role and the second role;
[0021] The instruction artificial intelligence model, referring to the description information of the first character, the description information of the second character, the target intimacy level, and the dialogue context data, simulates the identity of the first character and generates a target dialogue to be spoken to the second character, including:
[0022] The artificial intelligence model is instructed to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, the relationship information, and the dialogue context data.
[0023] Optionally, the information processing method further includes:
[0024] Obtain tone type information, which is used to describe the tone type of the speech;
[0025] Obtain relationship information, which describes the relationship between the first role and the second role;
[0026] The instruction artificial intelligence model, referring to the description information of the first character, the description information of the second character, the target intimacy level, and the dialogue context data, simulates the identity of the first character and generates a target dialogue to be spoken to the second character, including:
[0027] The artificial intelligence model is instructed to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, the tone type information, the relationship information and the dialogue context data.
[0028] Optionally, the target intimacy level is an intimacy level within a preset intimacy range;
[0029] The preset intimacy range includes multiple intimacy levels, with different intimacy levels representing different intensities of intimacy.
[0030] Optionally, the target intimacy level is used to indicate the degree of intimacy in conversation.
[0031] Optionally, the dialogue context data includes dialogue records and / or dialogue scenario information;
[0032] The dialogue record is the conversation between the first character and the second character.
[0033] Optionally, the description information of the first character and the description information of the second character include name, zodiac sign, and hometown information.
[0034] To achieve this objective, the embodiments of the present invention also employ the following technical solutions:
[0035] An information processing device, comprising:
[0036] The first acquisition unit is used to acquire the description information of the first role and the description information of the second role;
[0037] The second acquisition unit is used to acquire dialogue context data;
[0038] The generation unit is used to instruct the artificial intelligence model to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, and the dialogue context data.
[0039] Optionally, the information processing device further includes a third acquisition unit;
[0040] The third acquisition unit is used to acquire the target intimacy level, which represents the degree of intimacy.
[0041] The generation unit is also used to instruct the artificial intelligence model to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy and the dialogue context data.
[0042] The target dialogue corresponds to the level of intimacy described by the target intimacy.
[0043] Optionally, the information processing device further includes a fourth acquisition unit;
[0044] The fourth acquisition unit is used to acquire tone type information, which is used to describe the tone type of the speech.
[0045] The generation unit is further configured to instruct an artificial intelligence model to simulate the identity of the first character by referring to the description information of the first character, the description information of the second character, the target intimacy, the tone type information, and the dialogue context data, and to generate a target dialogue to be spoken to the second character in the tone type described by the tone type information.
[0046] Optionally, the information processing device further includes a fifth acquisition unit;
[0047] The fifth acquisition unit is used to acquire relationship information, which describes the relationship between the first role and the second role.
[0048] The generation unit is further configured to instruct an artificial intelligence model to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, the relationship information and the dialogue context data.
[0049] Optionally, the information processing device further includes a fourth acquisition unit and a fifth acquisition unit;
[0050] The fourth acquisition unit is used to acquire tone type information, which is used to describe the tone type of the speech.
[0051] The fifth acquisition unit is used to acquire relationship information, which describes the relationship between the first role and the second role.
[0052] The generation unit is also used to instruct an artificial intelligence model to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, the tone type information, the relationship information and the dialogue context data.
[0053] Optionally, the target intimacy level is an intimacy level within a preset intimacy range;
[0054] The preset intimacy range includes multiple intimacy levels, with different intimacy levels representing different intensities of intimacy.
[0055] Optionally, the target intimacy level is used to indicate the degree of intimacy in conversation.
[0056] Optionally, the dialogue context data includes dialogue records and / or dialogue scenario information;
[0057] The dialogue record is the conversation between the first character and the second character.
[0058] To achieve this objective, the embodiments of the present invention also employ the following technical solutions:
[0059] A computer device includes a processor and a memory, the memory storing a computer program that is loaded and executed by the processor to implement the information processing method described above.
[0060] To achieve this objective, the embodiments of the present invention also employ the following technical solutions:
[0061] A computer-readable storage medium storing a computer program, which is loaded and executed by a processor to implement the information processing method described above.
[0062] To achieve this objective, the embodiments of the present invention also employ the following technical solutions:
[0063] A computer program product includes computer instructions stored in a computer-readable storage medium, wherein a processor retrieves the computer instructions from the computer-readable storage medium, causing the processor to load and execute the instructions to implement the information processing method described above.
[0064] As can be seen from the above technical solutions, the embodiments of the present invention have the following advantages:
[0065] In the information processing method of this invention, obtaining descriptions of the first and second characters allows for understanding the information of both parties in the dialogue; obtaining dialogue context data allows for understanding the context of the dialogue. Then, an artificial intelligence model is instructed to simulate the identity of the first character, generate a target dialogue addressed to the second character, referencing the descriptions of the first and second characters and the dialogue context data. Because the artificial intelligence model can understand the information of both parties in the dialogue and the context of the dialogue, the generated target dialogue is more accurate and better meets user needs. Attached Figure Description
[0066] Figure 1 A flowchart illustrating an information processing method provided in an embodiment of the present invention;
[0067] Figure 2A flowchart of an information processing method provided in another embodiment of the present invention;
[0068] Figure 3 This is a schematic diagram of an information processing device according to another embodiment of the present invention. Detailed Implementation
[0069] This invention provides an information processing method, an information processing apparatus, and a computer device for generating dialogues that better meet user needs.
[0070] Figure 1 A flowchart of an information processing method provided in an embodiment of the present invention is shown below. Figure 1 The information processing method of this invention includes:
[0071] Step 101: Obtain the description information of the first character and the description information of the second character.
[0072] The description information of the first character is used to describe the first character, and the description information of the second character is used to describe the second character. By using the description information of the first character and the description information of the second character, one can understand the first character and the second character.
[0073] Step 102: Obtain dialogue context data.
[0074] Dialogue context data is used to represent contextual information about the dialogue.
[0075] Step 103: Instruct the artificial intelligence model to simulate the identity of the first character, generate the target dialogue to be spoken to the second character, referring to the description information of the first character, the description information of the second character, and the dialogue context data.
[0076] After obtaining the description information of the first character and the second character, as well as the dialogue context data, the artificial intelligence model can be instructed to simulate the identity of the first character and generate the target dialogue to be spoken to the second character, referring to the description information of the first character, the description information of the second character, and the dialogue context data.
[0077] Because the description of the first character describes the situation of the first character, the artificial intelligence model can simulate the identity of the first character by referring to the description of the first character. After referring to the description of the second character and the dialogue context data, the generated target dialogue can conform to the context of the dialogue context data and is also suitable for communicating with the second character. Therefore, the target dialogue is more accurate and can better meet the user's needs.
[0078] In summary, in the information processing method of this invention, obtaining the description information of the first role and the second role allows us to understand the information of both parties in the dialogue; obtaining the dialogue context data allows us to understand the context of the dialogue. Then, the artificial intelligence model is instructed to simulate the identity of the first role, generate a target dialogue for the second role, referencing the description information of the first role, the description information of the second role, and the dialogue context data. Because the artificial intelligence model can understand the information of both parties in the dialogue and the context of the dialogue, the generated target dialogue is more accurate and better meets user needs.
[0079] Figure 2 A flowchart of an information processing method provided in another embodiment of the present invention. Figure 2 The information processing method of the illustrated embodiment can be based on Figure 1 The information processing method of the illustrated embodiment is implemented. Wherein, Figure 2 The information processing method of the illustrated embodiment can be applied to smart terminals, such as mobile phones, tablets, computers, and laptops; it can also be used in data processing systems, such as a system composed of a mobile phone and a server; or it can be applied to a server. This embodiment of the invention does not specifically limit its application in this regard. The following description uses the application of the information processing method to a computer device as an example. The computer device can be a mobile phone, a server, or a combination of a mobile phone and a server. It should be understood that the application of the information processing method of this embodiment to a computer device is merely illustrative and does not constitute a limitation on the embodiments of the invention.
[0080] See Figure 2 The information processing method of this invention includes:
[0081] Step 201: Obtain the description information of the first character and the description information of the second character.
[0082] The computer device acquires description information of a first role and description information of a second role. The description information of the first role is used to describe the first role, and the description information of the second role is used to describe the second role.
[0083] The specific content of the descriptions for both the first and second roles can vary. Optionally, the descriptions for both roles may include name, zodiac sign, and hometown information. For example, the description for the first role could be: Name: Zhang San, Zodiac Sign: Aquarius, Hometown: Guangzhou. The description for the second role could be: Name: Li Si, Zodiac Sign: Pisces, Hometown: Shanghai.
[0084] It should be understood that the description information of the first character and the description information of the second character can also be implemented in other ways, such as describing the character's profession, describing the character's personality and hobbies, etc., and the embodiments of the present invention do not specifically limit this.
[0085] Step 202: Obtain dialogue context data.
[0086] Acquire dialogue context data, which represents the contextual information of the dialogue, including its linguistic context. There are several ways to implement dialogue context data, as detailed below:
[0087] In some embodiments, the dialogue context data includes a dialogue record, which is a conversation between a first role and a second role. There are several ways to obtain the dialogue record; it can be that the user inputs the conversation between the first and second roles, or the device recognizes a chat screenshot to obtain the conversation between the first and second roles.
[0088] In some embodiments, dialogue context data includes dialogue scenario information. Dialogue scenario information describes the context of the dialogue, such as saying good morning, saying good night, or comforting the other person. It can be understood that dialogue scenario information includes the context of the conversation, within which users can have a direction in their dialogue.
[0089] In some embodiments, the dialogue context data includes dialogue records and dialogue scenario information. The dialogue records are the conversation between a first character and a second character. In other words, the dialogue context data also includes information from the foregoing embodiments, such as the conversation between a first character and a second character in a scenario of comforting the other party.
[0090] Step 203: Obtain the target intimacy level.
[0091] Target intimacy is used to represent the degree of intimacy between characters. For example, target intimacy can be neutral or intimate. Target intimacy can be obtained through user input, system presets, or system data analysis.
[0092] The degree of intimacy can include the intimacy of the relationship or the intimacy of the tone of voice; however, this embodiment of the invention does not specifically limit this.
[0093] Optionally, target intimacy is used to indicate the level of intimacy in speech. In a conversational context, the level of intimacy can be reflected through the choice of words used in the conversation; for example, ordinary intimacy often uses polite language, while intimate intimacy often uses emotionally charged words.
[0094] Optionally, the target intimacy level is one of the intimacy levels within a preset intimacy range. The preset intimacy range includes multiple intimacy levels, with different levels representing different intensities of intimacy. For example, there may be 10 or 100 intimacy levels, with smaller values indicating greater unfamiliarity and larger values indicating greater intimacy.
[0095] Step 204: Instruct the artificial intelligence model to simulate the identity of the first character, generate the target dialogue to be spoken to the second character, referring to the description information of the first character, the description information of the second character, the target intimacy and dialogue context data.
[0096] Among them, the target dialogue and the target intimacy level correspond to the degree of intimacy described.
[0097] After obtaining the description information of the first character, the description information of the second character, the target intimacy, and the dialogue context data, the computer device can send the description information of the first character, the description information of the second character, the target intimacy, and the dialogue context data to the artificial intelligence model, instructing the artificial intelligence model to simulate the identity of the first character and generate the target dialogue to be spoken to the second character, thereby obtaining the target dialogue.
[0098] Because the description information of the first character is used to describe the first character, the artificial intelligence model can simulate the identity of the first character by referring to the description information of the first character, so that the target dialogue looks like it was spoken by the first character, or is consistent with the information of the first character. For example, if the description information of the first character includes that his hometown is Guangzhou, then the target dialogue may include "My hometown Guangzhou is very beautiful, let's go play together".
[0099] The description information of the second character is used to describe the second character. Referring to the description information of the second character allows the artificial intelligence model to refer to the description information of the second character when generating the target dialogue. For example, if the description information of the second character includes that the zodiac sign is Pisces, then the target dialogue may include "I heard that people born under the sign of Pisces are very emotional".
[0100] Target intimacy is used to represent the degree of intimacy. Referring to the target intimacy level, when an AI model generates a target dialogue, it can ensure that the vocabulary used in the target dialogue matches the corresponding intimacy level; that is, the target dialogue corresponds to the intimacy level described by the target intimacy level. In some embodiments, target intimacy is used to represent the degree of intimacy in speech. For example, an intimate intimacy level would involve more use of interjections and emoticons, and more words expressing a warm atmosphere. For example, a target dialogue could include "I miss you so much~".
[0101] Dialogue context data represents the contextual information of a dialogue, and the target dialogue can conform to the context expressed by the dialogue context data. For example, dialogue context data includes dialogue records, that is, the dialogue between the first character and the second character. The target dialogue can continue the dialogue based on these dialogue records. For example, if the second character says in the dialogue record, "I'm so tired today," the target dialogue could be "Then rest well at home after get off work," thereby helping the first character continue the dialogue.
[0102] Alternatively, if the dialogue context data includes dialogue scene information, the AI model can generate a target dialogue that matches the scene information. For example, if the dialogue scene information is "Good morning greeting", the target dialogue could include "Good morning, may today be another energetic day".
[0103] Alternatively, if the dialogue context data includes dialogue records and dialogue scene information, then the target dialogue can simultaneously conform to the context of both the dialogue records and the dialogue scene information. For example, if the dialogue scene information is "comforting the other party," and in the dialogue record, the second character says, "I haven't done my job well lately, and my boss scolded me, boohoo," based on this information, the AI model generates the target dialogue, which could be, "It's normal to make mistakes at work. Just do better next time. More importantly, you need to get enough rest so you have the energy to work."
[0104] It should be understood that the above descriptions often separate the reference to the descriptions of the first character, the second character, the target intimacy level, and the dialogue context data. However, the information processing method of this embodiment can instruct the artificial intelligence model to simultaneously reference these information, allowing the generated target dialogue to comprehensively consider the descriptions of the first character, the second character, the target intimacy level, and the dialogue context data. This makes the target dialogue more realistic and better meets the user's needs. It is understood that the specific method of referencing this information can be freely determined by the artificial intelligence model or specifically prompted by the system; this embodiment does not impose specific limitations on this.
[0105] Once the target conversation is obtained, it can be sent to a second role, or the user can copy the target conversation from the device for use.
[0106] Step 204 enables the instruction AI model to simulate the identity of the first character and generate the target dialogue for the second character by referring to the description information of the first character, the description information of the second character, and the dialogue context data. Of course, there are other implementation methods, which will not be elaborated here.
[0107] To ensure that the generated target dialogue more accurately meets the user's needs, the information processing method of this invention can be implemented in other ways, such as by incorporating more reference information. Please refer to the following description for details.
[0108] In some embodiments, the information processing method of this invention further includes: acquiring tone type information. Tone type information describes the tone type of speech. When users converse, their words carry tone, and tone type information can indicate a specific tone of voice, such as an emotional tone or a persona-based tone. For example, tone type information could be a tone of praise, a mature and composed tone, or a gentle tone.
[0109] Accordingly, step 204 is specifically implemented by instructing the artificial intelligence model to simulate the identity of the first character by referring to the description information of the first character, the description information of the second character, the target intimacy, the tone type information and the dialogue context data, and generating the target dialogue to be spoken to the second character in the tone type described by the tone type information.
[0110] In this way, the target dialogue can carry the tone described by tone type information. By controlling the tone type information, the target dialogue generated by the artificial intelligence model can meet more scenarios, because the tone required for dialogue is different in different scenarios. For example, a gentle tone is needed when comforting someone, and a mature and steady tone is needed when giving advice.
[0111] It should be understood that target intimacy is used to represent the degree of intimacy. Intimacy level and tone type information are not the same; even with the same tone type information describing the tone type, the degree of intimacy can vary. For example, when using a gentle tone, the stronger the intimacy level indicated by the target intimacy, the richer and more intimate the emotions expressed in the dialogue. Through the combined effect of target intimacy and tone type information, control over the target dialogue can achieve more granular control, thus meeting more scenarios and more accurately expressing user needs. In particular, in some examples, when target intimacy is used to represent the degree of intimacy in speech, and tone type information describes the tone type of speech, both are direct controls on language expression. This allows AI models to more accurately understand the needs of expression, resulting in more precise control over the word choice in the target dialogue.
[0112] In some embodiments, the information processing method of this invention further includes: obtaining relationship information. The relationship information describes the relationship between the first role and the second role; for example, the relationship information may be that of relatives, colleagues, lovers, or classmates.
[0113] Accordingly, step 204 is specifically implemented by instructing the artificial intelligence model to simulate the identity of the first character, generate the target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, the relationship information and the dialogue context data.
[0114] In this way, the target dialogue can correspond to relationship information. After understanding the relationship between the first and second roles, the AI model can generate a target dialogue that conforms to that relationship information. This allows the relationship information to influence the wording and expression of the target dialogue, achieving more accurate dialogue generation. For example, the wording will differ between romantic relationships and colleague relationships. By combining target intimacy and relationship information, the control over the target dialogue can reach a more granular level, thus meeting more scenarios and more accurately expressing user needs.
[0115] In some embodiments, the information processing method may further acquire tone type information, which describes the tone type of the speech; and acquire relationship information, which describes the relationship between the first character and the second character. In this case, step 204 is specifically implemented by instructing an artificial intelligence model to simulate the identity of the first character and generate a target dialogue for the second character, referencing the description information of the first character, the description information of the second character, the target intimacy, the tone type information, the relationship information, and the dialogue context data.
[0116] For a more detailed explanation of tone type and relationship information, please refer to the above text. Tone type information describes the tone of voice used in speaking, relationship information describes the relationship between the first and second characters, and target intimacy indicates the degree of intimacy. These three pieces of information describe different dimensions that should be considered in a dialogue. By referring to this information, the AI model can achieve more granular control and expression of the target dialogue, thus meeting the needs of more dialogue scenarios. For example, if the relationship information is colleagues and the tone type information is a gentle tone, different target intimacy levels will result in different word choices in the target dialogue. Closer colleagues will express themselves more casually, while lower-level colleagues can use a lower level of target intimacy, in which case the target dialogue will use more honorifics and more formal expressions. Alternatively, if the relationship information is lovers and the tone type information is a gentle tone, different target intimacy levels will result in different levels of intimacy in the target dialogue. For example, when lovers are discussing future plans or making up after an argument, they can use different levels of target intimacy to allow the AI model to generate corresponding target dialogues.
[0117] The artificial intelligence model in this embodiment of the invention can be a large language model (LLM) or a multimodal artificial intelligence model, such as one that can recognize images and input text content, or other artificial intelligence algorithms. This embodiment of the invention does not specifically limit the specific model.
[0118] The following section uses examples of large language models or multimodal AI models to illustrate how to instruct AI models. Instructions can be given to the AI model using prompts. The prompts include descriptions of the first and second characters, target intimacy, tone, relationship, and dialogue context data, allowing the AI model to reference them. Furthermore, the prompts should describe that the AI model should simulate the first character's identity, and specify that the task is to generate dialogue for the second character based on the context of the dialogue data.
[0119] For example, a prompt might look like this: "You are Zhang San, from Guangzhou. Please generate a message for Li Si based on the dialogue log. Please reply in a gentle tone, using a strong sense of intimacy.\n#Reference Information:\nZhang San's Information: Name: Zhang San, Zodiac Sign: Aquarius, Hometown: Guangzhou.\nLi Si's Information: Name: Li Si, Zodiac Sign: Pisces, Hometown: Shanghai.\nZhang San and Li Si are lovers.\n#Dialogue Log:\nZhang San: Have you eaten?\nLi Si: Not yet, I'm so hungry." Some AI models categorize prompts into system prompts and dialogue prompts. Dialogue logs can be written to either system prompts or dialogue prompts, while other information can be written to system prompts.
[0120] In summary, the information processing method of this invention involves obtaining descriptions of a first character and a second character to obtain information about them; acquiring dialogue context data to obtain contextual information about the dialogue; and acquiring a target intimacy level to represent the degree of intimacy. Then, an artificial intelligence model is instructed to simulate the identity of the first character, generate a target dialogue for the second character, referencing the descriptions of the first and second characters, the target intimacy level, and the dialogue context data. The target dialogue corresponds to the degree of intimacy described by the target intimacy level. By enabling the artificial intelligence model to understand the information of both parties and the required degree of intimacy during the dialogue, the target dialogue generated based on the contextual information of the dialogue context data offers greater control, better meets user needs, and more accurately satisfies users' dialogue expression needs in different scenarios.
[0121] Figure 3 This is a schematic diagram of an information processing device according to another embodiment of the present invention. Figure 3 The information processing apparatus of the illustrated embodiments can be used to execute the information processing methods shown in the above embodiments.
[0122] See Figure 3 The information processing apparatus of this invention includes:
[0123] The first acquisition unit 301 is used to acquire the description information of the first role and the description information of the second role;
[0124] The second acquisition unit 302 is used to acquire dialogue context data;
[0125] The generation unit 303 is used to instruct the artificial intelligence model to simulate the identity of the first character and generate the target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, and the dialogue context data.
[0126] Optionally, the information processing device further includes a third acquisition unit 304;
[0127] The third acquisition unit 304 is used to acquire the target intimacy level, which is used to represent the degree of intimacy.
[0128] The generation unit 303 is also used to instruct the artificial intelligence model to simulate the identity of the first character and generate the target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy and the dialogue context data.
[0129] Among them, the target dialogue and the target intimacy level correspond to the degree of intimacy described.
[0130] Optionally, the information processing device further includes a fourth acquisition unit 305;
[0131] The fourth acquisition unit 305 is used to acquire tone type information, which describes the tone type of the speech.
[0132] The generation unit 303 is also used to instruct the artificial intelligence model to simulate the identity of the first character by referring to the description information of the first character, the description information of the second character, the target intimacy, the tone type information and the dialogue context data, and to generate the target dialogue to be spoken to the second character in the tone type described by the tone type information.
[0133] Optionally, the information processing device further includes a fifth acquisition unit 306;
[0134] The fifth acquisition unit 306 is used to acquire relationship information, which describes the relationship between the first role and the second role;
[0135] The generation unit 303 is also used to instruct the artificial intelligence model to simulate the identity of the first character and generate the target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, the relationship information and the dialogue context data.
[0136] Optionally, the information processing device further includes a fourth acquisition unit 305 and a fifth acquisition unit 306;
[0137] The fourth acquisition unit 305 is used to acquire tone type information, which describes the tone type of the speech.
[0138] The fifth acquisition unit 306 is used to acquire relationship information, which describes the relationship between the first role and the second role;
[0139] The generation unit 303 is also used to instruct the artificial intelligence model to simulate the identity of the first character and generate the target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, the tone type information, the relationship information and the dialogue context data.
[0140] Optionally, the target intimacy level is one of the intimacy levels within a preset intimacy range;
[0141] The preset intimacy range includes multiple intimacy levels, with different intimacy levels representing different levels of intimacy.
[0142] Optionally, target intimacy is used to indicate the level of intimacy in conversation.
[0143] Optionally, the dialogue context data includes dialogue records and / or dialogue scenario information;
[0144] The dialogue log is a record of conversations between the first and second characters.
[0145] In summary, in the information processing apparatus of this embodiment, the first acquisition unit 301 acquires description information of the first character and the second character, thus understanding the information of both parties in the dialogue; the second acquisition unit 302 acquires dialogue context data, thus understanding the context of the dialogue; and the generation unit 303 instructs the artificial intelligence model to simulate the identity of the first character and generate a target dialogue for the second character, referencing the description information of the first character, the description information of the second character, and the dialogue context data. Because the artificial intelligence model can understand the information of both parties in the dialogue and the context of the dialogue, the generated target dialogue is more accurate and better meets user needs.
[0146] This invention also provides a computer device, which includes a processor and a memory. The memory stores a computer program, which is loaded and executed by the processor to implement the information processing methods described in the embodiments above.
[0147] This invention also provides a computer-readable storage medium storing a computer program that is loaded and executed by a processor to implement the information processing methods described in the embodiments above.
[0148] This invention also provides a computer program product, which includes computer instructions stored in a computer-readable storage medium. A processor retrieves the computer instructions from the computer-readable storage medium, causing the processor to load and execute them to implement the information processing methods described in the embodiments above.
[0149] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.
[0150] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.
[0151] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0152] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0153] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0154] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. An information processing method, characterized in that, include: Obtain the description information of the first character and the description information of the second character; Obtain dialogue context data; The artificial intelligence model is instructed to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, and the dialogue context data.
2. The information processing method according to claim 1, characterized in that, The information processing method further includes: Obtain the target intimacy level, which is used to represent the degree of intimacy; The instruction artificial intelligence model, referring to the description information of the first role, the description information of the second role, and the dialogue context data, simulates the identity of the first role and generates a target dialogue to be spoken to the second role, including: The artificial intelligence model is instructed to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, and the dialogue context data. The target dialogue corresponds to the level of intimacy described by the target intimacy.
3. The information processing method according to claim 2, characterized in that, The information processing method further includes: Obtain tone type information, which is used to describe the tone type of the speech; The instruction artificial intelligence model, referring to the description information of the first character, the description information of the second character, the target intimacy level, and the dialogue context data, simulates the identity of the first character and generates a target dialogue to be spoken to the second character, including: The artificial intelligence model is instructed to simulate the identity of the first character by referring to the description information of the first character, the description information of the second character, the target intimacy, the tone type information, and the dialogue context data, and to generate a target dialogue to be spoken to the second character using the tone type described by the tone type information.
4. The information processing method according to claim 2, characterized in that, The information processing method further includes: Obtain relationship information, which describes the relationship between the first role and the second role; The instruction artificial intelligence model, referring to the description information of the first character, the description information of the second character, the target intimacy level, and the dialogue context data, simulates the identity of the first character and generates a target dialogue to be spoken to the second character, including: The artificial intelligence model is instructed to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, the relationship information, and the dialogue context data.
5. The information processing method according to claim 2, characterized in that, The information processing method further includes: Obtain tone type information, which is used to describe the tone type of the speech; Obtain relationship information, which describes the relationship between the first role and the second role; The instruction artificial intelligence model, referring to the description information of the first character, the description information of the second character, the target intimacy level, and the dialogue context data, simulates the identity of the first character and generates a target dialogue to be spoken to the second character, including: The artificial intelligence model is instructed to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, the target intimacy, the tone type information, the relationship information and the dialogue context data.
6. The information processing method according to claim 2, characterized in that, The target intimacy level is an intimacy level within a preset intimacy range; The preset intimacy range includes multiple intimacy levels, with different intimacy levels representing different intensities of intimacy.
7. The information processing method according to claim 2, characterized in that, The target intimacy level is used to indicate the degree of intimacy in conversation.
8. The information processing method according to any one of claims 1 to 7, characterized in that, The dialogue context data includes dialogue records and / or dialogue scenario information; The dialogue record is the conversation between the first character and the second character.
9. An information processing device, characterized in that, include: The first acquisition unit is used to acquire the description information of the first role and the description information of the second role; The second acquisition unit is used to acquire dialogue context data; The generation unit is used to instruct the artificial intelligence model to simulate the identity of the first character and generate a target dialogue to be spoken to the second character, by referring to the description information of the first character, the description information of the second character, and the dialogue context data.
10. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing a computer program, the computer program being loaded and executed by the processor to implement the information processing method as described in any one of claims 1 to 8.