Activation management of multiple voice assistants

By activating the server to manage the multi-assistant platform, the problem of accidental activation on the multi-assistant platform is solved, ensuring that other assistants are not activated when one voice assistant is active, thus improving the user experience and the accuracy of interaction.

CN117542360BActive Publication Date: 2026-11-17MICROSOFT TECHNOLOGY LICENSING LLC
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311513748.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2019-05-22
Filing Date
2020-03-26
Publication Date
2026-11-17
Estimated Expiration
2040-03-26

AI Technical Summary

Technical Problem

On multi-assistant platforms, existing technologies struggle to effectively manage the activation of voice assistants, leading to issues such as environmental noise or false activation.

Method used

By activating a server to manage multiple voice assistants, it ensures that other voice assistants are not activated while one voice assistant is active, and employs keyword detection and status check mechanisms to prevent accidental activation.

Benefits of technology

It improves the user experience, prevents random activation of the voice assistant due to environmental noise or accidental activation, and protects the continuity and accuracy of user interaction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117542360B_ABST
    Figure CN117542360B_ABST
Patent Text Reader

Abstract

Activation management of multiple voice assistants is provided. A system and method include: detecting a user utterance of a voice keyword from a plurality of voice keywords, each of the plurality of voice keywords being associated with a corresponding voice assistant application; in response to the detection of a voice keyword from the plurality of corresponding voice keywords, determining whether any of the plurality of voice assistant applications is designated as uninterruptible; and if it is determined that none of the plurality of voice assistant applications is designated as uninterruptible, activating one of the plurality of voice assistant applications associated with the detected keyword.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] This application is a divisional application of the original Chinese invention patent application entitled "Activation Management of Multiple Voice Assistants", application number 202080037174.9, and application date March 26, 2020. Background Technology

[0002] Voice assistants are rapidly becoming ubiquitous in modern computing devices. Users can interact with voice assistants to control the functionality of computing devices. When a computing device receives and recognizes voice keywords associated with a voice assistant, the voice assistant is activated and launched to engage the user in a voice-driven computing experience.

[0003] Various computing platforms offer their own corresponding voice assistants. Each voice assistant is activated in response to a specific keyword, which can consist of one or more words. For example, on Windows operating systems, the Cortana voice assistant is activated with the keyword "Hey Cortana," on Google devices with the keyword "OK Google," and on Amazon devices with the keyword "Alexa."

[0004] The computing platform can be configured to support more than one voice assistant. Therefore, such a platform must be able to detect and respond to each keyword from each supported voice assistant. Furthermore, the system must manage activation when keywords from one or more assistants are received while one voice assistant is active. A system that effectively addresses the difficulties of managing voice assistant activation on a multi-assistant platform is desired. Attached Figure Description

[0005] Figure 1 This is a block diagram of a multi-assistant platform during operation according to some embodiments;

[0006] Figure 2 This is a block diagram of a multi-assistant platform during operation according to some embodiments;

[0007] Figure 3 This is a block diagram of a multi-assistant platform during operation according to some embodiments;

[0008] Figure 4 This is a block diagram of a multi-assistant platform during operation according to some embodiments;

[0009] Figure 5 This is a block diagram of a multi-assistant platform during operation according to some embodiments;

[0010] Figure 6 This is a block diagram of a multi-assistant platform during operation according to some embodiments;

[0011] Figure 7 This is a block diagram of a multi-assistant platform during operation according to some embodiments;

[0012] Figure 8 This is a block diagram of a multi-assistant platform during operation according to some embodiments;

[0013] Figure 9 This is a block diagram of a multi-assistant platform during operation according to some embodiments;

[0014] Figure 10 This is a block diagram of a multi-assistant platform during operation according to some embodiments;

[0015] Figure 11 This is a flowchart illustrating the process of managing voice assistant activation on a multi-voice assistant platform according to some embodiments;

[0016] Figure 12 This is a flowchart illustrating the process of managing voice assistant activation on a multi-voice assistant platform according to some embodiments;

[0017] Figure 13 This is a block diagram of the software components of a multi-voice assistant platform according to some embodiments;

[0018] Figure 14 The illustration shows a device that can support multiple voice assistant platforms according to some embodiments;

[0019] Figure 15 This is a block diagram of a cloud computing system supporting multiple voice assistants according to some embodiments; and

[0020] Figure 16 This is a block diagram of a computing system implementing a multi-voice assistant platform according to some embodiments. Detailed Implementation

[0021] The following description is provided to enable anyone in the art to make and use the described embodiments. However, various modifications will still be apparent to those skilled in the art.

[0022] The embodiments described herein provide technical solutions to the problem of managing the activation of multiple voice assistants provided by a single computing platform. According to some embodiments, only one voice assistant is active at a given time. If a voice keyword of another voice assistant is detected while the first voice assistant is active, the other voice assistants are not activated, and the active voice assistant is not interrupted.

[0023] Some embodiments thus protect the user experience from random activation due to ambient noise or the detection of keywords used in the current conversation with the assistant. In the latter case, an example is considered where the user says, “Hello Cortana, remind me to call my wife Alexa.” According to some embodiments, the keyword “Hello Cortana” causes activation of the Cortana voice assistant, while the keyword “Alexa” does not cause activation of the Alexa voice assistant. Details of several alternative implementations are provided below.

[0024] Figure 1 This is a block diagram of a multi-assistant platform 100 during operation according to some embodiments. Platform 100 consists of an activation server 110, a voice assistant 120, and a voice assistant 130. Typically, a user 140 speaks a keyword 150, and the activation server 110 detects the keyword 150 and activates one of the voice assistants 120 and 130 based on the keyword 150 and the logic described herein.

[0025] Platform 100 can be implemented by any suitable one or more computing devices. In some embodiments, platform 100 is a personal computer running an operating system, and activation server 110 is a component thereof. Voice assistants 120 and 130 may include applications compatible with the operating system and capable of communicating with activation server 110 via various application programming interfaces (APIs). Each component described herein can be implemented by executable code.

[0026] As described below, platform 100 can be implemented in other devices running an operating system, such as smartphones or smart speakers. One or more of activation server 110, voice assistant 120, and voice assistant 130 can be located away from other components of platform 100, for example, in a cloud-based deployment.

[0027] One or more of the activation server 110, voice assistant 120, and voice assistant 130 can be implemented in the device firmware. Two or more of the activation server 110, voice assistant 120, and voice assistant 130 can be provided by different software vendors.

[0028] In this example, Figure 1The dashed lines indicate that voice assistants 120 and 130 are initially inactive when user 140 speaks keyword 150. For the purposes of this example, keyword 150 is a voice activation keyword associated with voice assistant 120. According to some embodiments, the audio signal representing keyword 150 is received by an audio input device (e.g., a microphone) of platform 110, and other components of platform 110 process the audio signal to identify keyword 150, determine that keyword 150 is associated with voice assistant 120, and transmit a corresponding message to activation server 110. In some embodiments, other components process the audio signal to identify keyword 150, determine that keyword 150 is associated with a registered voice assistant, and pass keyword 150 to activation server 110. Activation server 110 then determines that keyword 150 is associated with voice assistant 120.

[0029] As mentioned above, each of voice assistants 120 and 130 can be associated with a corresponding dedicated keyword, and a "keyword" can consist of one or more words. In some embodiments, one or both of voice assistants 120 and 130 can be associated with more than one corresponding keyword, and each keyword can include more than one word. Some embodiments are envisioned in which two or more voice assistants share the same one or more keywords.

[0030] Regardless of how the activation server 110 determines that the voice keywords of the voice assistant 120 have been received, such as Figure 2 As shown, activation server 110 transmits an activation signal to voice assistant 120. Subsequent activation of voice assistant 120 is indicated by a thick solid line. As will be described below, voice assistant 120 may perform a second check on the received audio signal before activation to confirm that keyword 150 has been received (i.e., the received audio signal represents keyword 150).

[0031] like Figure 3 As depicted, the active voice assistant 120 can send a signal that the voice assistant 120 is uninterruptible (e.g., using an API exposed by the activation server 110). Therefore, the activation server 110 sets a flag or stores other data indicating that the voice assistant 120 is uninterruptible.

[0032] The conversation then takes place between the active voice assistant 120 and the user 140, as is known in the art. For simplicity, Figure 4This dialogue session is illustrated as occurring directly between user 140 and voice assistant 120; however, according to known systems, multiple hardware and software components can typically facilitate the dialogue session. In some examples, the dialogue session may require voice assistant 120 to access other components of platform 100 (e.g., “Set an alarm for 6 a.m.”) and / or external services (e.g., “What is today’s weather forecast?”), details of which are omitted in this document for clarity.

[0033] Figure 5 The illustration shows user 140 uttering keyword 160 during an ongoing conversation with active voice assistant 120. For example, suppose voice assistant 120 is associated with the keyword "Hello Xiaona" and voice assistant 130 is associated with the keyword Alexa. Figures 1 to 5 It can be described that the platform 100 operates while saying "Hello Xiaona, calling my wife Alexa".

[0034] like Figure 6 As shown, even after receiving the keyword 160 associated with voice assistant 130, voice assistant 120 remains active and engages in a conversational session with user 140. As will be described in detail below, in some embodiments, speaking keyword 160 causes activation server 110 to receive the instruction for keyword 160. In response, activation server 110 determines that voice assistant 120 is active and has been marked as uninterruptible. Therefore, activation server 110 does not send an activation signal to voice assistant 130, even though platform 100 receives and detects the keyword 160 associated with voice assistant 130.

[0035] Figure 7 The termination of a conversational session between user 140 and voice assistant 120 is depicted. Specifically, voice assistant 120 transmits a signal to activation server 110 indicating that voice assistant 120 will become inactive. In some embodiments, upon entering the final speaking state of the session, voice assistant 120 sends a signal to activation server indicating that voice assistant 120 is no longer uninterruptible (e.g., an "interruptible" signal), and then... Figure 7 As depicted, it becomes inactive. This embodiment allows another voice assistant to be activated before voice assistant 120 becomes inactive.

[0036] Figure 8 The state of platform 100 is depicted after voice assistant 120 indicates to activation server 110 that it is inactive. Voice assistants 120 and 130 are inactive, and no conversation session with user 140 is in progress.

[0037] Now suppose user 140 says keyword 160, such as Figure 9As shown, keyword 160 is a voice activation keyword associated with voice assistant 130. Similarly, the audio signal representing keyword 160 can be received by the audio input device of platform 110, and other components of platform 110 (not shown) can process the audio signal to identify keyword 160, determine that keyword 160 is associated with voice assistant 130, and transmit the corresponding message to activation server 110. Alternatively, other components process the audio signal to identify keyword 160, determine that keyword 160 is associated with a registered voice assistant, and transmit keyword 160 to activation server 110. Activation server 110 then determines that the received keyword 160 is associated with voice assistant 130.

[0038] In response to determining that the voice keywords of the voice assistant 160 have been received, the activation server 110 transmits an activation signal to the voice assistant 130, such as... Figure 10 As shown. Voice assistant 130 receives an activation signal and activates in response. Voice assistant 130 can also perform a second check on the received audio signal before activation to confirm that the received audio signal represents keyword 160.

[0039] Figure 11 This is a flowchart of process 1100 according to some embodiments. Process 1100 and other processes described herein can be performed using any suitable combination of hardware and software. The software program code implementing these processes can be stored by any non-transient tangible medium, including fixed disks, volatile or non-volatile random access memory, DVDs, flash drives, or magnetic tapes, and executed by any number of processing units, including but not limited to processors, processor cores, and processor threads. Embodiments are not limited to the examples described below.

[0040] Process 1100 may be executed by an activation server, such as activation server 110 of platform 100, but the embodiments are not limited thereto. Process 1100 may be executed by a platform that supports two or more voice assistants.

[0041] The process loops in S1110 until a keyword associated with the voice assistant is detected. One or more voice assistants in the platform execution process 1100 may be associated with two or more keywords. Detecting keywords in S1110 may include receiving a keyword detection signal from another component of the platform. Keyword detection in S1110 may include receiving an audio signal (e.g., via passive monitoring of the surrounding audio environment) and determining that the received audio signal represents a keyword associated with a voice assistant registered to the platform. This determination may include speech-to-text processing of the audio signal and comparing the resulting text with a list of keywords associated with the registered voice assistant.

[0042] Once a keyword associated with the voice assistant is detected, the process proceeds from S1110 to S1120. In S1120, it is determined whether any registered voice assistants are currently uninterruptible. As described above, a previously activated voice assistant can indicate itself as uninterruptible, for example, using an API exposed by the activation server. This indication can be stored and used in the determination in S1120.

[0043] According to some embodiments, the uninterruptibility of the voice assistant is determined based on the voice assistant's state. For example, the voice assistant may be in an inactive or active state. Active states may include states such as speaking, listening, and thinking. In S1120, since the voice assistant is in a "thinking" active state, it can be determined that the platform's voice assistant is uninterruptible.

[0044] The determination in S1120 can be based on any suitable combination of voice assistant status information and / or uninterruptibility flags.

[0045] If it is determined in S1120 that any currently registered voice assistant is uninterruptible, no activation is performed in response to the detected keyword, and the process returns to S1110 to wait for the keyword detection described above.

[0046] If it is determined that the unregistered voice assistant is uninterruptible, the process proceeds from S1120 to S1130. In S1130, activation is sent to the voice assistant associated with the detected keyword. The process then returns to S1110 to await keyword detection. Meanwhile, the voice assistant that received the activation can activate and begin a conversation with the user. As described above, for any subsequent determinations in S1120, the voice assistant can also indicate that it is uninterruptible.

[0047] Figure 12 The diagram illustrates an alternative process 1200 that can be executed by an activation server of a multi-voice assistant platform. Process 1200 can be executed to manage the activation of the voice assistant based on received keywords.

[0048] Initially, in S1210, it is determined whether a keyword associated with the registered voice assistant has been detected. The process loops in S1210 until such a keyword is detected. As mentioned above, keyword detection may include, for example, receiving a keyword detection signal from another component of the platform, or receiving an audio signal and determining that the received audio signal represents a keyword associated with the voice assistant registered to the platform.

[0049] Once a keyword associated with the voice assistant is detected, it is determined in S1220 whether the associated voice assistant is active. This determination can be based on the associated voice assistant's reports and monitoring status. If the associated voice assistant is not active, the process proceeds to S1230 to determine whether the platform's voice assistant is currently marked as uninterruptible.

[0050] The assessment of a voice assistant's uninterruptibility can be based on whether a flag indicating that the voice assistant is uninterruptible is currently set. This flag can be set in response to a previous uninterruptibility indication received from the voice assistant. The determination of uninterruptibility can also, or alternatively, be based on the voice assistant's current state. According to some embodiments, a voice assistant that has marked itself as uninterruptible but then becomes inactive is no longer considered uninterruptible. In some embodiments, the voice assistant needs to indicate that it is interruptible before becoming inactive.

[0051] If it is determined in S1230 that the unregistered voice assistant is uninterruptible, activation is sent to the voice assistant associated with the detected keyword. The voice assistant is then activated and enters a dialogue session known in the art. The activated voice assistant can indicate itself as uninterruptible during the dialogue session. Meanwhile, the process returns to S1210 to await keyword detection.

[0052] Returning to S1220, if the voice assistant associated with the detected keyword is active, the process proceeds to S1240 to send an activation signal to the voice assistant. This action allows the voice assistant to determine whether to interrupt its current conversation session based on whether it has marked itself as uninterruptible after detecting its own keyword.

[0053] The implementation may support any other desired logic for interrupting or not interrupting an active voice assistant in response to the detection of associated keywords or keywords associated with other voice assistants.

[0054] Figure 13 The illustration shows the software architecture 1300 of a computing system implementing an embodiment. This embodiment may include, in addition to... Figure 13 Software components beyond those illustrated. Software architecture 1300 can be implemented on any suitable computing platform, including, for example, personal computers, smartphones, and IoT edge devices running operating systems such as smart speakers. Software architecture 1300, or portions thereof, can be implemented within embedded systems. One or more components of software architecture 1300 can be deployed on virtual machines on cloud-based servers. Each component of software architecture 1300 can be implemented using executable code.

[0055] The Activation Runtime User Service 1310 may include operating system components. The Activation Runtime User Service 1310 is responsible for activating voice assistant applications registered within the software architecture 1300 when an activation event is detected.

[0056] For example, microphone 1311 can receive audio signals, and keyword detection component 1312 can detect registered voice keywords represented within the audio signals. The detection is exposed to the activation runtime user service 1310 via voice activation manager 1313.

[0057] The Activation Runtime User Service 1310 can detect other activation events, such as button presses via Human-Machine Interface Device Service 1314. Activation events received via Bluetooth can also be exposed as button presses via Bluetooth Service 1315. Any suitable logic can be implemented to determine whether to send activation to the voice assistant application based on the button press activation event. In some embodiments, the Activation Runtime User Service 1310 sends activation to the corresponding voice assistant in response to a received button press activation event, regardless of the activity or inactivity state of any active voice assistant or whether any voice assistant has been marked as uninterruptible. Given the degree of intent that can be inferred from a button press, this embodiment is intended to reflect user intent.

[0058] According to some embodiments, the conversation manager 1316 may execute process 1100 or process 1200 to determine whether to send an activation to the voice assistant application based on a received activation event. In this example, it is assumed that a voice keyword associated with the voice assistant application 1320 is received when neither the voice assistant application 1320 nor the voice assistant application 1330 is activated. The activation manager 1317 then sends an activation to the voice assistant application 1320 to activate the voice trigger 1322, which is a background task associated with the voice assistant application 1320.

[0059] In response to activation of the voice trigger 1322, the voice assistant application 1320 verifies that its keyword has been received. In this regard, the voice assistant application 1320 includes an assistant-specific keyword detector 1324. The assistant-specific keyword detector 1324 performs verification upon receiving activation to confirm that the associated keyword has been spoken. This verification allows the detection component 1312 to implement a less robust detection model than necessary, thereby saving platform resources.

[0060] If authentication is successful, a conversational client 1326 is created. The conversational client 1326 can then provide its status and interruptibility information to the conversational manager 1316. For example, the conversational client 1326 can indicate that it is active, listening, and uninterruptible. The conversational manager 1316 maintains a record of this information.

[0061] Therefore, upon receiving another keyword detection, the conversation session manager 1316 can determine whether to send activation to the associated voice assistant application based on the status and / or uninterruptibility information of all voice assistant applications in the software architecture 1300. This determination can follow the logic described above with respect to processes 1100 and 1200 and / or any other suitable logic. If the active voice assistant has not marked itself as interruptible (e.g., via communication between the conversation session client 1326 and the conversation session manager 1316), it can be determined that activation should simply be sent to the corresponding voice assistant as soon as the keyword is detected.

[0062] Figure 14 The illustration shows a computing device that can implement a system according to some embodiments. Figure 14 The computing devices do not include a non-exhaustive set of systems that can be implemented in the embodiments.

[0063] Each of devices 1410 to 1450 can run an operating system including the activation manager described above. The operating system may also allow registration of more than one voice assistant application. Personal computer 1410 and laptop computer 1420 can run a desktop operating system, while smartphone 1440 and smartwatch 1450 can run a mobile operating system. Smart speaker 1430 can run a mobile operating system or a dedicated smart device operating system. Again, the embodiments are not limited to these descriptions.

[0064] Figure 15 The diagram illustrates a distributed system 1500 according to some embodiments. The distributed system 1500 can provide multiple voice assistant functionality to in-vehicle devices that do not currently possess such functionality. The cloud-based components of system 1500 can be implemented using on-demand virtual machines, virtual servers, and cloud storage instances.

[0065] according to Figure 15 For example, client devices 1510 and 1520 may selectively connect to cloud service 1530 to transmit detected keywords. In some embodiments, one or both of client devices 1510 and 1520 send all received audio signals to cloud service 1530 in order to receive a voice assistant service based on any keywords that happen to appear in the received audio signals.

[0066] Assistant activation service 1532 can be implemented as a cloud service that provides activation of a voice assistant application based on received audio signals or keyword detection events. Cloud service 1532 may include a voice assistant application 1534 that can be activated by assistant activation service 1532. Cloud service 1550, which may be operated by a service provider different from the service provider operating cloud service 1530, can provide voice assistant application 1552. Voice assistant application 1552 can also be activated by assistant activation service 1532 based on signals received from client devices 1510 and / or 1520.

[0067] Each of the voice assistant applications 1534 and 1552 can communicate with the assistant activation service 1532 to mark itself as interruptible. Therefore, the assistant activation service 1532 can operate as described herein to selectively determine whether to send activation to either the voice assistant application 1534 or the voice assistant application 1552.

[0068] Figure 16 This is a block diagram based on some examples of System 1600. System 1600 may include any suitable computing system and may execute program code to provide the voice assistant described herein.

[0069] System 1600 includes a processing unit 1610 operatively coupled to a communication device 1620, a persistent data storage system 1630, one or more input devices 1640, one or more output devices 1650, and volatile memory 1660. The processing unit 1610 may include one or more processors, processing cores, etc., for executing program code. The communication interface 1620 facilitates communication with an external network. The input devices 1640 may include, for example, a keyboard, keypad, mouse or other pointing device, microphone, touchscreen, and / or eye-tracking device. The output devices 1650 may include, for example, a display and / or speakers.

[0070] Data storage system 1630 may include any number of suitable persistent storage devices, including combinations of magnetic storage devices (such as magnetic tape, hard disk drives, and flash memory), optical storage devices, read-only memory (ROM) devices, etc. Memory 1660 may include random access memory (RAM), storage class memory (SCM), or any other fast access memory.

[0071] Voice assistant application 1632 may include program code executed by processing unit 1610 to cause system 1600 to perform a conversational session with a user. Each voice assistant application in voice assistant application 1632 is associated with a voice keyword used for its activation, as described herein. Operating system 1634 includes operating system components known in the art and may also include activation services to selectively activate the voice assistant applications 1632 described herein. Data 1636 may include any data used by any system or application of system 1600. Such data may include session state information and / or uninterruptible flags described herein.

[0072] Each functional component described herein can be implemented, at least in part, in computer hardware, program code, and / or in one or more computing systems that execute such program code known in the art. Such a computing system may include one or more processing units that execute processor-executable program code stored in a memory system.

[0073] The foregoing figures illustrate a logical architecture for describing the process according to some embodiments, and actual implementations may include more or different components arranged in other ways. Other topologies may be used in conjunction with other embodiments. Moreover, each component or device described herein may be implemented by any number of devices communicating via any number of other public and / or private networks. Two or more such computing devices may be geographically isolated from each other and may communicate with each other via any number of networks and / or dedicated connections of any known manner. Each component or device may include any number of hardware and / or software elements suitable for providing the functionality described herein as well as any other functionality. For example, any computing device used in an implementation of a system according to some embodiments may include a processor for executing program code such that the computing device operates as described herein.

[0074] All systems and processes discussed herein can be implemented in program code stored on one or more non-transient computer-readable media. Such media may include, for example, hard disks, DVD-ROMs, flash drives, magnetic tapes, and solid-state random access memory (RAM) or read-only memory (ROM) storage units. Therefore, embodiments are not limited to any specific combination of hardware and software.

[0075] Those skilled in the art will understand that various adaptations and modifications of the above embodiments can be configured without departing from the claims. Therefore, it is to be understood that the claims may be practiced in ways different from those specifically described herein.

Claims

1. A computing system comprising one or more processors, said one or more processors being configured to execute processor-executable program code stored on a non-transient medium, such that the computing system: activating a first voice assistant application to perform a first user dialog session, the first voice assistant application being associated with the first voice keyword; as well as While the first voice assistant application is active and the first user conversation is in progress: Receive a second audio signal, the second audio signal representing a second voice keyword associated with a second voice assistant application; In response to the reception of the second audio signal, it is determined that the first voice assistant application is uninterruptible, wherein the second voice assistant application remains inactive in response to the reception of the second audio signal; Receive a signal from the first voice assistant application indicating that the first voice assistant application can be interrupted; Receive a third audio signal, wherein the third audio signal represents the second voice keyword; as well as In response to the reception of the third audio signal, it is determined that the first voice assistant application can be interrupted, and an activation signal is transmitted to the second voice assistant application to activate the second voice assistant application and execute the second user dialogue session.

2. The computing system according to claim 1, wherein the first voice assistant application is provided by a first cloud service provider, and the second voice assistant application is provided by a second cloud service provider.

3. The computing system according to claim 2, comprising a microphone for receiving a first audio signal, a second audio signal, and a third audio signal representing the first voice keyword.

4. The computing system of claim 1, wherein the computing system is provided by a first cloud service provider.

5. The computing system of claim 4, wherein the first voice assistant application is provided by the first cloud service provider, and the second voice assistant application is provided by the second cloud service provider.

6. The computing system of claim 1, comprising one or more processing units for executing processor-executable program code, such that the computing system: While the second voice assistant application is active and the second user conversation session is being executed: Receive a second signal from the second voice assistant application, indicating that the second voice assistant application should not be interrupted; Receive a fourth audio signal, wherein the fourth audio signal represents the first voice keyword; In response to the reception of the fourth audio signal, it is determined that the second voice assistant application is uninterruptible, wherein the first voice assistant application remains inactive in response to the reception of the fourth audio signal; Receive a third signal from the second voice assistant application indicating that the second voice assistant application can be interrupted; Receive a fifth audio signal, wherein the fifth audio signal represents the first voice keyword; as well as In response to the reception of the fifth audio signal, it is determined that the second voice assistant application can be interrupted, and an activation signal is transmitted to the first voice assistant application to activate the first voice assistant application and execute the third user dialogue session.

7. The computing system of claim 6, wherein the computing system is provided by a first cloud service provider, the first voice assistant application is provided by the first cloud service provider, and the second voice assistant application is provided by a second cloud service provider.

8. A method for a computing device, comprising: An activation signal is transmitted to a first voice assistant application to activate the first voice assistant application and execute a first user dialogue session, the first voice assistant application being associated with a first voice keyword; as well as While the first voice assistant application is active and the first user conversation is in progress: Receive a second audio signal, the second audio signal representing a second voice keyword associated with a second voice assistant application; In response to the reception of the second audio signal, it is determined that the first voice assistant application is uninterruptible, wherein the second voice assistant application remains inactive in response to the reception of the second audio signal; Receive a first signal from the first voice assistant application indicating that the first voice assistant application can be interrupted; Receive a third audio signal, wherein the third audio signal represents the second voice keyword; as well as In response to receiving the third audio signal, it is determined that the first voice assistant application can be interrupted, and an activation signal is transmitted to the second voice assistant application to activate the second voice assistant application and execute the second user dialogue session.

9. The method of claim 8, wherein the first voice assistant application is provided by a first cloud service provider, and the second voice assistant application is provided by a second cloud service provider.

10. The method of claim 9, wherein the first audio signal, the second audio signal, and the third audio signal representing the first voice keyword are received by the microphone of the computing device.

11. The method of claim 8, wherein the computing device is provided by a first cloud service provider.

12. The method of claim 11, wherein the first voice assistant application is provided by the first cloud service provider, and the second voice assistant application is provided by the second cloud service provider.

13. The method of claim 8, further comprising: While the second voice assistant application is active and the second user conversation session is being executed: Receive a second signal from the second voice assistant application, indicating that the second voice assistant application should not be interrupted; Receive a fourth audio signal, wherein the fourth audio signal represents the first voice keyword; In response to the reception of the fourth audio signal, it is determined that the second voice assistant application is uninterruptible, wherein the first voice assistant application remains inactive in response to the reception of the fourth audio signal; Receive a third signal from the second voice assistant application indicating that the second voice assistant application can be interrupted; Receive a fifth audio signal, wherein the fifth audio signal represents the first voice keyword; as well as In response to the reception of the fifth audio signal, it is determined that the second voice assistant application can be interrupted, and an activation signal is transmitted to the first voice assistant application to activate the first voice assistant application and execute the third user dialogue session.

14. The method of claim 13, wherein the computing device is provided by a first cloud service provider, the first voice assistant application is provided by the first cloud service provider, and the second voice assistant application is provided by a second cloud service provider.

15. A system comprising: Multiple voice assistant applications, the multiple voice assistant applications being executed by at least a processor, a first voice assistant application among the multiple voice assistant applications being associated with a first voice keyword, and a second voice assistant application among the multiple voice assistant applications being associated with a second voice keyword; The voice assistant activation component, executed at least by the processor, to: In response to the detection of the first voice keyword, an activation signal is transmitted to the first voice assistant application to activate the first voice assistant application and execute the first user dialogue session; While the first voice assistant application is active and the first user conversation is in progress: Receive a second user speech, which represents the second voice keyword; In response to the reception of the second user's voice, it is determined that the first voice assistant application is uninterruptible, wherein the second voice assistant application remains inactive in response to the reception of the second user's voice; Receive a signal from the first voice assistant application indicating that the first voice assistant application can be interrupted; Receive a third user's speech, wherein the third user's speech represents the second voice keyword; as well as In response to the reception of the third user's speech, it is determined that the first voice assistant application is interruptible, and an activation signal is transmitted to the second voice assistant application to activate the second voice assistant application and execute the second user dialogue session.

16. The system of claim 15, wherein the first voice assistant application is provided by a first cloud service provider, and the second voice assistant application is provided by a second cloud service provider.

17. The system of claim 16, wherein the first user utterance, the second user utterance, and the third user utterance representing the first voice keyword are received by the microphone of the system.

18. The system of claim 15, wherein the voice assistant activation component is provided by a first cloud service provider.

19. The system of claim 18, wherein the first voice assistant application is provided by the first cloud service provider, and the second voice assistant application is provided by the second cloud service provider.

20. The system of claim 15, wherein the voice assistant activation component is used to: While the second voice assistant application is active and the second user conversation session is being executed: Receive a second signal from the second voice assistant application, indicating that the second voice assistant application should not be interrupted; Receive a fourth user utterance, wherein the fourth user utterance represents the first voice keyword; In response to the reception of the fourth user's speech, it is determined that the second voice assistant application is uninterruptible, wherein the first voice assistant application remains inactive in response to the reception of the fourth user's speech; Receive a third signal from the second voice assistant application indicating that the second voice assistant application can be interrupted; Receive a fifth user utterance, wherein the fifth user utterance represents the first voice keyword; as well as In response to the reception of the fifth user's speech, it is determined that the second voice assistant application is interruptible, and an activation signal is transmitted to the first voice assistant application to activate the first voice assistant application and execute the third user dialogue session.

Citation Information

Patent Citations

  • Control method and apparatus of multiple voice assistants, and computer readable storage medium

    CN107919123A

  • Context-based user agent

    CN109196464A