CN-114375449-B - Techniques for dialog processing using context data

Abstract

Techniques are described for improving the efficiency and accuracy of dialog processing tasks using data stored in association with a context level for a user. The dialog system stores historical dialog data associated with a plurality of configured context levels. The dialog system receives the utterance and recognizes terms for disambiguation from the utterance. Based on the determined context level, the dialog system identifies relevant historical data stored to the database. The historical data may be used to perform tasks such as resolving ambiguities based on user preferences, disambiguating named entities based on previous dialogs, and identifying previously generated answers to queries. Based on the context level, the dialog system can effectively identify relevant information and use the identified information to provide a response.

Inventors

• M.E. Johnson

Assignees

• 甲骨文国际公司
• 甲骨文国际公司

Dates

Publication Date

20260421

Application Date

20200827

Priority Date

20200826

Claims (20)

1. 1. A method for context-based dialog processing, comprising: Receiving, by the dialog system, a spoken utterance from a user; processing, by the dialog system, the spoken utterance to identify terms for disambiguation; determining, by the dialog system, a context level of a term for disambiguation among a plurality of predefined context levels, wherein the plurality of predefined context levels includes an immediate context, a short-term context, a mid-term context, and a long-term context; Identifying, by the dialog system, a value of the term based on the determined context level using a database storing a plurality of values of the user in association with the context level, and The identified value is used by the dialog system to disambiguate the term.
2. 2. The method of claim 1, further comprising: generating, by the dialog system, a logical form of the spoken utterance based on the disambiguated terms; generating a response by the dialog system based on the logical form, and Outputting, by the dialog system, the response to the user.
3. 3. The method of claim 2, further comprising: identifying by the dialog system from the database the stored execution results based on the logical form, Wherein the response is also generated based on the execution result.
4. 4. A method according to claim 2 or 3, further comprising: the representation of the spoken utterance, the logical form, and the response are stored by the dialog system to the database.
5. 5. A method according to any one of claims 1 to 3, wherein identifying the value comprises: Selecting a database for the context level from a plurality of context-level based databases based on the determined context level, and The selected database is queried using the user's identifier and the term to identify the value.
6. 6. A method according to any one of claims 1 to 3, further comprising: identifying, by the dialog system, stored data for a plurality of previous dialogues with the user from the database, and The identified data is analyzed to calculate the value.
7. 7. A non-transitory computer-readable memory storing a plurality of instructions executable by one or more processors, the plurality of instructions comprising instructions that when executed by the one or more processors cause the one or more processors to perform a process comprising: Receiving a spoken utterance from a user; Processing the spoken utterance to identify terms for disambiguation; Determining a context level of a term for disambiguation among a plurality of predefined context levels, wherein the plurality of predefined context levels includes an immediate context, a short-term context, a mid-term context, and a long-term context; identifying a value of the term based on the determined context level using a database storing a plurality of values of the user in association with the context level, and The identified value is used to disambiguate the term.
8. 8. The non-transitory computer readable memory of claim 7, the process further comprising: generating a logical form of the spoken utterance based on the disambiguated terms; Preparing a response based on the logical form, and Outputting the response to the user.
9. 9. The non-transitory computer readable memory of claim 8, the process further comprising: Identifying stored execution results from the database based on the logical form, Wherein the response is also generated based on the execution result.
10. 10. The non-transitory computer readable memory of claim 8 or 9, the process further comprising: storing a representation of the spoken utterance, the logical form, and the response to the database.
11. 11. The non-transitory computer readable memory of any one of claims 7 to 9, wherein identifying the value comprises: Selecting a database for the context level from a plurality of context-level based databases based on the determined context level, and The selected database is queried using the user's identifier and the term to identify the value.
12. 12. The non-transitory computer readable memory of any one of claims 7 to 9, the process further comprising: identifying stored data from the database for a plurality of previous conversations with the user, and The identified data is analyzed to calculate the value.
13. 13. A system for context-based dialog processing, comprising: One or more processors; A memory coupled to the one or more processors, the memory storing a plurality of instructions for execution by the one or more processors, the plurality of instructions including instructions that when executed by the one or more processors cause the one or more processors to perform a process comprising: Receiving a spoken utterance from a user; Processing the spoken utterance to identify terms for disambiguation; Determining a context level of the term for disambiguation among a plurality of predefined context levels, wherein the plurality of predefined context levels includes an immediate context, a short-term context, a mid-term context, and a long-term context; identifying a value of the term based on the determined context level using a database storing a plurality of values of the user in association with the context level, and The identified value is used to disambiguate the term.
14. 14. The system of claim 13, the process further comprising: generating a logical form of the spoken utterance based on the disambiguated terms; Preparing a response based on the logical form, and Outputting the response to the user.
15. 15. The system of claim 14, the process further comprising: Identifying stored execution results from the database based on the logical form, Wherein the response is also generated based on the execution result.
16. 16. The system of claim 14 or 15, the process further comprising: The representation of the spoken utterance, the logical form, and the response are stored to a database.
17. 17. The system of any of claims 13 to 15, wherein identifying the value comprises: Selecting a database for the context level from a plurality of context-level based databases based on the determined context level, and The selected database is queried using the user's identifier and the term to identify the value.
18. 18. The system of any of claims 13 to 15, the process further comprising: identifying stored data from the database for a plurality of previous conversations with the user, and The identified data is analyzed to calculate the value.
19. 19. A dialog system, comprising: means for receiving a spoken utterance from a user; means for processing the spoken utterance to identify terms for disambiguation; Means for determining a context level of the term for disambiguation among a plurality of predefined context levels, wherein the plurality of predefined context levels includes an immediate context, a short term context, a mid term context, and a long term context, Means for preparing a response in accordance with the disambiguated terms based on the context level, and Means for outputting the response to the user.
20. 20. The dialog system of claim 19, further comprising: Means for identifying a value of the term based on the determined context level using a database, the database storing a plurality of values of the user in association with the context level; Means for disambiguating the term using the identified value; means for generating a logical form of the spoken utterance based on the disambiguated terms, and Means for preparing the response based on the logical form.

Description

Techniques for dialog processing using context data Cross Reference to Related Applications Under 35 U.S. C.119 (e), the present application claims the benefits and priorities of U.S. application Ser. No. 62/899,649 entitled "CONTEXT-BASED DIALOG TECHNIQUES" (CONTEXT based dialog technology) filed on at 9, 12, 2019 and U.S. application Ser. No. 17/003,250 filed on 26, 8, 2020, the contents of which are incorporated herein by reference in their entirety for all purposes. Technical Field The present disclosure relates generally to dialog systems. More specifically, but not by way of limitation, the present disclosure describes techniques for using historical context levels to affect conversational tasks such as named entity linking and result ordering. Background Nowadays, more and more devices enable users to interact directly with the device using speech or spoken words. For example, a user may speak with such a device in natural language, where the user may ask questions or make a statement requesting to perform certain actions. In response, the device performs the requested operation or answers the user's question using voice output. Because direct use of speech for interaction is a more natural and intuitive way for humans to communicate with the surrounding environment, the popularity of such speech-based systems is growing at astronomical digital speeds. Current dialog systems have limited ability to utilize context. Systems exist that discern context information based on recent inputs. For example, some existing systems may parse the pronouns by identifying the person who is the roll-call in the last sentence received by the system. However, in conventional systems, the context information is not fully utilized. Disclosure of Invention The present disclosure relates generally to dialog systems. More particularly, techniques are described for using historical context levels to affect conversational tasks such as named entity linking and result ordering. Various embodiments are described herein, including methods, systems, non-transitory computer-readable storage media storing programs, code, or instructions executable by one or more processors, and the like. In some embodiments, the dialog system receives a spoken utterance from a user. The dialog system processes the spoken utterance to identify terms for disambiguation. The dialog system determines a context level for the disambiguated term from a plurality of predefined context levels. Based on the determined context level, the dialog system identifies a value for the term using a database storing a plurality of values for the user in association with the context level and disambiguates the term using the identified value. In some aspects, the dialog system also generates a logical form for the spoken utterance based on the disambiguated terms, generates a response based on the logical form, and outputs the response to the user. In some aspects, the dialog system further identifies stored execution results from the database based on the logical form, wherein the response is further generated based on the execution results. In some aspects, the dialog system also stores representations, logical forms, and responses of spoken utterances to a database. In some aspects, identifying the value includes selecting a database for the context level from a plurality of context-level based databases based on the determined context level, and querying the selected database with an identifier and terminology of the user to identify the value. In some aspects, the dialog system also identifies stored data for a plurality of previous dialogs with the user from the database and analyzes the identified data to calculate the value. In some aspects, the plurality of predefined context levels includes an immediate context, a short-term context, a mid-term context, and a long-term context. Embodiments also include systems and computer-readable media (e.g., non-transitory computer-readable memory) configured to perform the methods described herein. The above described and other features and embodiments will become more apparent with reference to the following description and drawings. Drawings Fig. 1 is a simplified block diagram illustrating a dialog system in accordance with some embodiments. Fig. 2 is a simplified block diagram illustrating a context level database of the dialog system of fig. 1, in accordance with some embodiments. Fig. 3 is a simplified flow diagram depicting a method for context-based dialog processing, in accordance with some embodiments. FIG. 4 is a simplified flow diagram depicting additional context-based dialog processing techniques in accordance with certain embodiments. Fig. 5 is a simplified flowchart depicting a method for generating context-based values that may be used in the processes of fig. 3-4, in accordance with certain embodiments. Fig. 6 depicts a simplified diagram of a distributed system for implementing an embodiment. FIG. 7 is