Voice interaction method, computer program product and system for configuring engine according to user information

Abstract

The invention provides a voice interaction method, a computer program product and a system for configuring an engine according to user information, and the method is applied to a voice interaction system of a telephone voice assistant, carries out voice interaction with an incoming call user in a call process, recognizes a fraud risk according to the interaction content, and prevents the user from being fraud. The voice interaction system comprises a plurality of voice recognition engines, a plurality of big language model engines and a plurality of character-to-voice engines, the voice recognition engines, the big language model engines and the character-to-voice engines are in different levels, and the method comprises the following steps executed before voice interaction: obtaining user information, and recognizing an interaction demand level of a user from the user information; according to the interaction demand level of the user, the voice recognition engine, the large language model engine and the text-to-voice engine of the corresponding level are selected for system configuration, a voice interaction system corresponding to the user level is formed, and the personalized demand of the user is met.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a voice interaction method, a computer program product and a system for configuring an engine according to user information. Background Art

[0002] With the rapid development of artificial intelligence technology, voice interaction systems based on artificial intelligence have emerged. Telephone voice assistants are a common scenario for voice interaction systems. After being turned on, telephone voice assistants can answer calls on behalf of users, interact with callers during calls, and identify fraud risks based on the content of the interaction to prevent users from being defrauded.

[0003] A voice interaction system usually includes a speech recognition engine, a large language model engine, and a text-to-speech engine. Existing voice interaction systems generally configure these engines before use and cannot be personalized for different users. Summary of the invention

[0004] The technical problem to be solved by the present invention is how to enable the voice interaction system to be personalized for different users.

[0005] In order to solve the above technical problems, the present invention provides a voice interaction method for configuring an engine according to user information. The method is applied to a voice interaction system of a telephone voice assistant. The voice interaction system includes a voice recognition engine, a large language model engine, and a text-to-speech engine. The voice recognition engine, the large language model engine, and the text-to-speech engine each have multiple different levels. The method includes the following steps performed before voice interaction:

[0006] A. Obtain user information and identify the user's interaction demand level;

[0007] B. Select the corresponding level of speech recognition engine, large language model engine and text-to-speech engine for system configuration according to the user's interaction demand level.

[0008] Furthermore, in step A, identifying the user's interaction requirement level specifically includes: acquiring identity information input by the user, and identifying the user's interaction requirement level through the identity information.

[0009] Furthermore, the step B specifically includes: according to the user's interaction demand level, selecting one from multiple speech recognition engines at different levels, selecting one from multiple large language model engines at different levels, and selecting one from multiple text-to-speech engines at different levels, and then connecting the selected speech recognition engine, large language model engine and text-to-speech engine to achieve hot-swappable characteristics.

[0010] The present invention also provides a computer program product, including a computer program, which implements the steps in the above-mentioned voice interaction method when executed by a processor.

[0011] The present invention also provides a voice interaction system that configures an engine according to user information, which is applied to a telephone voice assistant, including a voice recognition engine, a large language model engine and a text-to-speech engine, wherein the voice recognition engine, the large language model engine and the text-to-speech engine each have multiple different levels, and also includes a main control module that is respectively connected to each voice recognition engine, each large language model engine and each text-to-speech engine, the main control module includes a memory and a processor that are interconnected, and the memory stores the computer program product as described above.

[0012] Furthermore, it also includes a voice activity detector, which is connected to the voice recognition engine.

[0013] The present invention has the following beneficial effects: before voice interaction, user information is first obtained to identify the user's interaction demand level, and then a speech recognition engine, a large prediction model engine and a text-to-speech engine of corresponding levels are selected according to the user's interaction demand level to perform system configuration, thereby forming a voice interaction system corresponding to the user level to meet the user's personalized needs. DETAILED DESCRIPTION

[0014] The present invention is further described in detail below in conjunction with specific implementation methods.

[0015] The present embodiment provides a voice interaction system that configures an engine according to user information. The voice interaction system is applied to a telephone voice assistant, including a voice activity detector (VAD), a voice recognition engine (ASR), a large language model engine (LLM), a text-to-speech engine (TTS), and a main control module, wherein the voice recognition engine, the large language model engine, and the text-to-speech engine each have multiple different levels, the voice activity detector is used to detect voice sentence input in real time, the voice recognition engine is used to convert the voice sentence into voice text, the large language model engine is used to generate a response text according to the voice text, and the text-to-speech engine is used to convert the response text into a response voice. The main control module is respectively connected to the voice activity detector, each voice recognition engine, each large language model engine, and each text-to-speech engine, and the main control module includes a memory and a processor connected to each other, and a computer program product is stored in the memory, and the computer program product includes a computer program. After the phone voice assistant is turned on, it can answer the phone on behalf of the user. During the call, the phone voice assistant interacts with the caller through the voice interaction system, identifies fraud risks based on the interaction content, and prevents the user from being defrauded. Before the voice interaction, the processor of the main control module executes the computer program in the memory. When the computer program is executed by the processor, a voice interaction method for configuring the engine according to user information is implemented, which specifically includes the following steps A and B implemented by the main control module.

[0016] A. Obtain user information and identify the user's interaction demand level from it. Specifically, before using the voice interaction system of the telephone voice assistant for voice interaction, obtain the identity information input by the user, and identify the user's interaction demand level through the identity information, where the interaction demand level is the service level that the user subscribes to, such as the VIP level of the telephone voice assistant service.

[0017] B. Select the corresponding level of speech recognition engine, large language model engine and text-to-speech engine for system configuration according to the user's interaction demand level. Specifically, after identifying the user's interaction demand level, the main control module selects one from multiple speech recognition engines of different levels, one from multiple large language model engines of different levels, and one from multiple text-to-speech engines of different levels, and then connects the selected speech recognition engine, large language model engine and text-to-speech engine with hot-swappable characteristics for system configuration, forming a speech interaction system corresponding to the user level to meet the user's personalized needs.

[0018] After configuring the speech recognition engine, large language model engine and text-to-speech engine, the main control module connects the speech activity detector to the speech recognition engine, uses the speech activity detector to detect the voice sentence input of the caller in real time, and sends the voice sentence to the speech recognition engine, then the speech recognition engine converts the voice sentence into voice text, the large language model engine generates a response text based on the voice text, and the text-to-speech engine converts the response text into a response voice, thereby realizing voice interaction. In the subsequent voice interaction process, if any of the speech recognition engine, large language model engine and text-to-speech engine fails, based on the hot-swappable feature connection between the three, other levels of the same type of engine can be selected for replacement connection to maintain the normal operation of the voice interaction system.

[0019] The above is only an implementation method of the invention, and does not limit the scope of patent protection. Those skilled in the art can make non-substantial changes or substitutions based on the invention, which still fall within the scope of patent protection.

Claims

1. A voice interaction method for configuring an engine based on user information, characterized in that: The method is applied to a voice interaction system of a telephone voice assistant, the voice interaction system includes a voice recognition engine, a large language model engine and a text-to-speech engine, wherein the voice recognition engine, the large language model engine and the text-to-speech engine each have multiple different levels, and the method includes the following steps performed before voice interaction: A. Obtain user information and identify the user's interaction demand level; B. Select the corresponding level of speech recognition engine, large language model engine and text-to-speech engine for system configuration according to the user's interaction demand level.

2. The voice interaction method according to claim 1, characterized in that: In the step A, the interaction requirement level of the user is identified by: obtaining identity information input by the user, and identifying the interaction requirement level of the user through the identity information.

3. The voice interaction method according to claim 1, characterized in that: Specifically, step B includes: selecting one from multiple speech recognition engines at different levels, selecting one from multiple large language model engines at different levels, and selecting one from multiple text-to-speech engines at different levels according to the user's interaction demand level, and then connecting the selected speech recognition engine, large language model engine, and text-to-speech engine with hot-swappable features.

4. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps in the voice interaction method described in any one of claims 1 to 3 are implemented.

5. A voice interaction system that configures an engine based on user information, which is applied to a telephone voice assistant, characterized in that: It includes a speech recognition engine, a large language model engine and a text-to-speech engine, wherein the speech recognition engine, the large language model engine and the text-to-speech engine are provided in multiples at different levels, and also includes a main control module connected to each speech recognition engine, each large language model engine and each text-to-speech engine respectively, the main control module includes a memory and a processor connected to each other, and the memory stores the computer program product according to claim 4.

6. The voice interaction system according to claim 5, further comprising a voice activity detector, wherein the voice activity detector is connected to the voice recognition engine.