WO2019227370A1 - 一种多语音助手控制方法、装置、系统及计算机可读存储介质 - Google Patents

一种多语音助手控制方法、装置、系统及计算机可读存储介质 Download PDF

Info

Publication number
WO2019227370A1
WO2019227370A1 PCT/CN2018/089166 CN2018089166W WO2019227370A1 WO 2019227370 A1 WO2019227370 A1 WO 2019227370A1 CN 2018089166 W CN2018089166 W CN 2018089166W WO 2019227370 A1 WO2019227370 A1 WO 2019227370A1
Authority
WO
WIPO (PCT)
Prior art keywords
wake
voice
external instruction
assistants
marker
Prior art date
Application number
PCT/CN2018/089166
Other languages
English (en)
French (fr)
Inventor
董学章
Original Assignee
Dong Xuezhang
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Dong Xuezhang filed Critical Dong Xuezhang
Priority to PCT/CN2018/089166 priority Critical patent/WO2019227370A1/zh
Priority to CN201880091294.XA priority patent/CN111954868A/zh
Publication of WO2019227370A1 publication Critical patent/WO2019227370A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt

Definitions

  • the present invention relates to the field of voice technology, and in particular, to a method, device, system, and computer-readable storage medium for controlling multiple voice assistants.
  • voice assistants can implement intelligent dialogues or instant Q & A with users to help users solve difficult problems.
  • existing electronic devices usually only have one voice assistant built-in. Even if multiple voice assistants are built in, some voice assistants may wake up when one voice assistant works, and multiple voice assistants working at the same time will cause conflicts (interference) ), The user experience is not ideal.
  • the purpose of the present invention is to provide a multi-voice assistant control method, device, system and computer-readable storage medium, which are used to solve the problems of conflicts (interference) caused by multiple voice assistants of existing electronic devices working simultaneously, and the user experience is not ideal. .
  • a first aspect provides a multi-voice assistant control method.
  • the method is applied to an electronic device that integrates multiple voice assistants.
  • the method steps include:
  • the plurality of voice assistants determine whether to perform an interrupt response process or wake up according to the external instruction, and the wake-up functions of the plurality of voice assistants are all in an enabled state.
  • the step "the plurality of voice assistants determine whether to perform an interrupt response process or wake up according to the external instruction” includes the steps:
  • the plurality of voice assistants input the external instruction
  • the plurality of voice assistants determine whether the external instructions are consistent with respective preset first wake-up tags
  • the corresponding voice assistant is woken up; otherwise, the response process of the corresponding voice assistant is interrupted.
  • the step "the plurality of voice assistants determine whether the external instruction is consistent with their respective preset first wake-up markers” includes the following steps:
  • the step determine whether the external instruction is consistent with a respective preset first wake-up marker.
  • the step "determining whether the external instruction includes a voice trigger event” includes the steps:
  • the external instruction includes a keyword associated with the identification of any one of the plurality of voice assistants, it is determined that the external instruction includes the voice trigger event.
  • the step "the plurality of voice assistants determine whether the external instruction is consistent with the respective preset first wake-up marker” includes the steps:
  • the step "the plurality of voice assistants determine whether to perform an interrupt response process or wake up according to the external instruction” includes the steps:
  • One voice assistant wakes up according to the external instruction determination, while other voice assistants perform interrupt response processes according to the external instruction determination.
  • the step "obtaining external instructions through the current multiple voice assistants of the electronic device” includes the following steps:
  • the step "the plurality of voice assistants determine whether to perform an interrupt response process or wake up according to the external instruction” includes the following steps:
  • the voice assistant that has determined to wake up sends the external instruction to the cloud server for processing through the network;
  • a second aspect of the present invention provides a multi-voice assistant control device.
  • the device is applied to an electronic device that integrates multiple voice assistants.
  • the device includes:
  • An acquisition module configured to acquire external instructions through multiple current voice assistants of the electronic device
  • a judging module is configured for the plurality of voice assistants to perform an interrupt response process or wake up according to the external instruction, wherein the wake-up functions of the plurality of voice assistants are in an enabled state.
  • the determination module includes:
  • a wake-up judging sub-module for the plurality of voice assistants to determine whether the external instruction is consistent with a respective preset first wake-up marker
  • a wake-up sub-module configured to wake up a corresponding voice assistant when the external instruction is consistent with the first wake-up marker
  • An interruption submodule is configured to interrupt a response process of a corresponding voice assistant when the external instruction is inconsistent with the first wake-up marker.
  • the determination module includes:
  • a trigger event judging sub-module configured to determine whether the external instruction includes a voice trigger event
  • the wake-up judging sub-module is configured to, when the external instruction includes the voice trigger event, the plurality of voice assistants determine whether the external instruction is consistent with a respective preset first wake-up marker.
  • the trigger event determination sub-module includes:
  • a keyword judging submodule configured to judge whether the external instruction includes a keyword associated with an identifier of any one of the plurality of voice assistants
  • a trigger event confirmation submodule is configured to determine that the external instruction includes the voice trigger event when the external instruction includes a keyword associated with an identifier of any one of the plurality of voice assistants.
  • the wake-up judging sub-module includes:
  • An association submodule configured to associate the keyword with a preset corresponding second wake-up marker when it is determined that the external instruction includes the voice trigger event
  • a marker judging sub-module for the plurality of voice assistants to determine whether the second wake-up marker is consistent with a respective preset first wake-up marker
  • the determination module includes:
  • the wake-up submodule is configured to wake up the corresponding voice assistant when the second wake-up marker is consistent with the respective preset first wake-up marker;
  • the interruption sub-module is configured to interrupt a response process of a corresponding voice assistant when the second wake-up marker is inconsistent with a respective preset first wake-up marker.
  • the device includes:
  • Microphone array acquisition module used to collect external input instructions through the microphone array.
  • a third aspect of the present invention provides a multi-voice assistant control system.
  • the system includes:
  • the multi-voice assistant control device includes:
  • a transceiver module configured to send an external instruction to the cloud server and / or receive a result processed by the cloud server through a voice assistant that has been determined to wake up;
  • the response module is configured to respond to the processed result of the cloud server received by the transceiver module through the voice assistant that has been determined to be woken up.
  • the invention also provides a multi-voice assistant control device, which is applied to an electronic device integrating multiple voice assistants, and includes:
  • a memory for storing the processor-executable instructions
  • the processor is configured to:
  • the plurality of voice assistants determine whether to perform an interrupt response process or wake up according to the external instruction, and the wake-up functions of the plurality of voice assistants are all in an enabled state.
  • the present invention also provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the steps of the method described in the first aspect.
  • a voice assistant wakes up according to instructions, and at the same time, other voice assistants interrupt the response process according to instructions, and multiple voice assistants will not be in the wake-up state at the same time, avoiding mutual interference and affecting user experience.
  • the preset wake-up word can be customized by the user, which not only enriches the use method of the voice assistant, but also avoids the single role of the voice assistant.
  • switching the wake-up assistant through voice command conversion also solves the technical problem that the existing electronic device needs to manually enable the voice assistant when switching the voice assistant, and the switching mode is not flexible enough.
  • the user can select a voice assistant from a plurality of voice assistants for question and answer based on his own preference and acceptance, which can be adapted to different users, and the human-computer interaction is more interesting.
  • the microphone array collects externally input instructions, which has a long pickup distance, good noise suppression, and better acquisition effect.
  • FIG. 1 is a schematic flowchart of a multi-voice assistant control method in a first embodiment of the present invention.
  • FIG. 2 is a schematic flowchart of a multi-voice assistant control method in a second embodiment of the present invention.
  • FIG. 3 is a schematic flowchart of a multi-voice assistant control method in a third embodiment of the present invention.
  • FIG. 4 is a schematic flowchart of a multi-voice assistant control method in a fourth embodiment of the present invention.
  • FIG. 5 is a schematic flowchart of a multi-voice assistant control method in a fifth embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of a multi-voice assistant control device in a sixth embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of a multi-voice assistant control device in a seventh embodiment of the present invention.
  • FIG. 8 is a schematic structural diagram of a multi-voice assistant control device in an eighth embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of a multi-voice assistant control device in a ninth embodiment of the present invention.
  • FIG. 10 is a schematic structural diagram of a multi-voice assistant control system in a tenth embodiment of the present invention.
  • the "response process” mentioned in this article refers to the audio process that responds (responses) according to the user's instructions after the voice assistant is woken up.
  • the response process and the audio process are used interchangeably.
  • the "monitoring state” mentioned in this article refers to the state in which the voice assistant waits for the activation of the wake-up word before being awakened, and the voice assistant in the monitoring state has no response process.
  • Working state The "working state” mentioned in this article refers to: the audio process of the voice assistant is active.
  • Re-awakening refers to: When the audio process of the voice assistant is active, the wake-up word is reactivated, and the audio process that was previously active is interrupted.
  • FIG. 1 is a flowchart of the method.
  • the method in this embodiment is applied to an electronic device that integrates multiple voice assistants.
  • step 100 external instructions are obtained through a plurality of current voice assistants of the electronic device.
  • step 101 the plurality of voice assistants determine whether to perform an interrupt response process or wake up according to the external instruction, wherein the wake-up functions of the plurality of voice assistants are in an enabled state.
  • step 101 includes steps:
  • One voice assistant wakes up according to the external instruction determination, while other voice assistants perform interrupt response processes according to the external instruction determination.
  • multiple current voice assistants of the electronic device obtain external instructions, and the external instructions may be voice information.
  • Multiple voice assistants perform interrupt response process or wake-up according to external command judgments, for example: a voice assistant is in a listening state (not awakened) and wakes up according to an external command, while a voice assistant is in a working state and performs an interrupt response process according to an external command; and For example: a voice assistant is in the working state to wake up (re-wake up) according to external instructions, while a voice assistant is in the monitoring state (not awakened) to perform an interrupt response process (to maintain the monitoring state) according to external instructions to ensure that it will not appear at the same time A situation where multiple voice assistants work at the same time resulting in conflict (interference).
  • FIG. 2 is a flowchart of the method.
  • the method in this embodiment is applied to an electronic device that integrates multiple voice assistants.
  • step 200 an externally input instruction is collected through the microphone array.
  • step 201 obtaining external instructions through a plurality of current voice assistants of the electronic device.
  • step 202 the plurality of voice assistants input the external instructions.
  • step 203 the plurality of voice assistants determine whether the external command is consistent with their respective preset first wake-up tags, and if so, execute step 204: wake up the corresponding voice assistant, otherwise execute step 205: interrupt the corresponding The response process of the voice assistant.
  • step 201 may implement the function of step 100
  • steps 202, 203, 204, and 205 may implement the function of step 101.
  • the microphone array can take a variety of forms: linear, ring, and spherical, for example: 2 microphone array, 6 + 1 microphone array, and 8 + 1 microphone array, with long pickup distance, good noise suppression, and better collection effect .
  • the third embodiment of the present invention relates to a multi-voice assistant control method
  • FIG. 3 is a flowchart of the method.
  • the method in this embodiment is applied to an electronic device that integrates multiple voice assistants.
  • step 300 an externally input instruction is collected through the microphone array.
  • step 301 obtaining external instructions through the multiple current voice assistants of the electronic device.
  • step 302 the plurality of voice assistants input the external instructions.
  • step 303 determine whether the external instruction includes keywords associated with any one of the plurality of voice assistants, and if so, execute step 304: the plurality of voice assistants determine whether the external instruction is related to The preset first wake-up tags are the same, otherwise, the above step 300 is performed.
  • Step 304 The plurality of voice assistants determine whether the external instruction is consistent with the respective preset first wake-up tag. If so, execute step 305: wake up the corresponding voice assistant, otherwise execute step 306: interrupt the corresponding voice assistant. Response process.
  • step 300 may implement the function of step 200
  • step 301 may implement the function of step 201
  • step 302 may implement the function of step 202
  • step 304 may implement the function of step 203
  • step 305 may implement the function of step 204
  • step 306 may The function of step 205 is implemented.
  • FIG. 4 is a flowchart of the method.
  • the method in this embodiment is applied to an electronic device that integrates multiple voice assistants.
  • step 400 an externally input instruction is collected through the microphone array.
  • step 401 obtaining external instructions through a plurality of current voice assistants of the electronic device.
  • step 402 the plurality of voice assistants input the external instructions.
  • step 403 determine whether the external instruction includes a keyword associated with any one of the plurality of voice assistant identifiers, and if so, execute step 404: compare the keyword with a preset corresponding second The wake-up tag is associated, otherwise step 400 described above is performed.
  • step 404 the process proceeds to step 405: the plurality of voice assistants determine whether the second wake-up marker is consistent with the respective preset first wake-up marker, and if so, execute step 406: wake up the corresponding voice assistant, otherwise execute Step 407: Interrupt the response process of the corresponding voice assistant.
  • step 400 may implement the function of step 300
  • step 401 may implement the function of step 301
  • step 402 may implement the function of step 302
  • step 403 may implement the function of step 303
  • steps 404 and 405 may implement the function of step 304
  • step 406 may implement the function of step 305
  • step 407 may implement the function of step 306.
  • multiple voice assistants are provided with corresponding wake-up words (keywords), and there is a preset correspondence relationship between the voice assistants and preset wake-up words (keywords), for example: wake-up of voice assistant A
  • the word is little A
  • the wake-up word of voice assistant B is little B
  • the wake-up word of voice assistant C is little C.
  • the voice assistant is Wake up, other voice assistants interrupt the response process of the corresponding voice assistant.
  • the microphone array collects an externally inputted instruction including "small A”.
  • Multiple voice assistants receive the instruction and determine that the instruction contains a wake-up word (keyword): "small A", and the wake-up word corresponds to a preset corresponding first
  • the two wake-up markers are associated (the second wake-up marker corresponding to small A is 00, the second wake-up marker corresponding to small B is 01, and the second wake-up marker corresponding to small C is 02).
  • the two wake-up markers (00) are compared with the respective preset first wake-up markers (the first wake-up marker preset by Voice Assistant A is 00, the first wake-up marker preset by Voice Assistant B is 01, and the voice Assistant C presets the first wake-up marker as 02), voice assistant A determines that the second wake-up marker (00) is the same as the preset first wake-up marker (00), and wakes up (if the voice assistant A was before If it is in the listening state, it will be woken up, and if it was in the previous working state, it will be woken up again.) Voice assistant B and voice assistant C determine the second wakeup marker (00) and the first wakeup marker (01) preset by themselves. , (02) are inconsistent, the response process of the corresponding voice assistant is interrupted (if Before sound assistant B / C is to monitor voice assistant state, it remains the listening state, if the previous work state, the interrupt response process corresponding voice assistant).
  • the wake-up word (keyword) of Baidu Voice Assistant is "small”
  • the wake-up word (Keyword) of Amazon Voice Assistant is "Alexa”.
  • Baidu Voice Assistant and Amazon Voice Assistant are integrated and applied to electronic devices. When the user first issued a "small” instruction to the electronic device, Baidu's voice assistant was woken up, and Amazon's voice assistant remained listening; after the user issued an "Alexa" instruction to the electronic device, Baidu's voice assistant interrupted the response process, and Amazon Voice The assistant was awakened.
  • the Baidu voice assistant when the user first issues a "small” instruction to the electronic device, the Baidu voice assistant is awakened, and the Amazon voice assistant remains listening; after the user issues a "small” instruction to the electronic device again, the Baidu voice assistant restarts Wake up, Amazon Voice Assistant remains listening.
  • the wake-up word (keyword) of the voice assistant may be the name of each voice assistant, or a word or phrase customized by the user according to actual needs. If there is no preset correspondence between the preset wake-up word and the voice assistant, the user-defined preset wake-up word does not specifically specify which voice assistant to wake up, the system can directly wake up the default voice assistant or prompt the user to choose which specific wake-up again A voice assistant.
  • voice switching can be performed between multiple voice assistants at any time, and multiple voice assistants will not be in the wake-up state at the same time, avoiding mutual interference and affecting the user experience.
  • the preset wake-up word can be customized by the user, which not only enriches the use of the voice assistant, avoids the singular role of the voice assistant, switches the wake-up assistant by voice command conversion, but also solves the need to manually switch the voice assistant for existing electronic devices.
  • Technical problems that the voice assistant is enabled and the switching mode is not flexible enough.
  • FIG. 5 is a flowchart of the method.
  • the method in this embodiment is applied to an electronic device that integrates multiple voice assistants.
  • step 500 external instructions are obtained through a plurality of current voice assistants of the electronic device.
  • step 501 the plurality of voice assistants determine whether to perform an interrupt response process or wake up according to the external instruction.
  • step 502 the voice assistant that has determined to wake up sends the external instruction to the cloud server for processing through the network.
  • step 503 responding to the processed result of the cloud server through the voice assistant that has been determined to wake up.
  • step 500 may implement the function of step 100
  • step 501 may implement the function of step 101.
  • the Baidu voice assistant and the Amazon voice assistant are integrated and applied to an electronic device (the wake-up word (keyword) of the Baidu voice assistant is "small", and the wake-up word (keyword) of the Amazon voice assistant is "Alexa").
  • Baidu Voice Assistant The user first issued a "small, Shanghai weather today" command to the electronic device, and Baidu Voice Assistant was woken up (Amazon Voice Assistant kept listening).
  • Baidu Voice Assistant uploaded the command to the cloud server via the network, and the cloud server processed and The result (which can be a voice packet) is sent back to Baidu Voice Assistant, and the Baidu Voice Assistant responds to the result of the process (sending "The weather in Shanghai today is cloudy, 25 °").
  • Amazon Voice Assistant was woken up (the audio / response process before Baidu Voice Assistant was interrupted), and Amazon Voice Assistant uploaded the command to the cloud server, cloud via the network
  • the server processes according to the instruction and sends the result (which can be a voice packet) back to the Amazon voice assistant, and the Amazon voice assistant responds to the result of the processing (issues "Today, weather, cloud, etc.”).
  • the user can select a voice assistant for question answering (QA) from a plurality of voice assistants according to his preferences and acceptance, which can be adapted to different users.
  • QA question answering
  • FIG. 6 is a schematic structural diagram of the device.
  • the device in this embodiment is applied to an electronic device that integrates multiple voice assistants.
  • the device includes:
  • An obtaining module 600 configured to obtain external instructions through a plurality of current voice assistants of the electronic device
  • the determining module 601 is configured to determine, by the multiple voice assistants, an interrupt response process or wake-up according to the external instruction, wherein the wake-up functions of the multiple voice assistants are in an enabled state.
  • the first embodiment is a method embodiment corresponding to this embodiment, and this embodiment can be implemented in cooperation with the first embodiment.
  • the related technical details mentioned in the first embodiment are still valid in this embodiment, and in order to reduce repetition, details are not repeated here. Accordingly, the related technical details mentioned in this embodiment can also be applied in the first embodiment.
  • a seventh embodiment of the present invention relates to a multi-voice assistant control device.
  • FIG. 7 is a schematic structural diagram of the device.
  • the device in this embodiment is applied to an electronic device that integrates multiple voice assistants.
  • This embodiment is improved based on the sixth embodiment. The improvements are as follows:
  • the device further includes:
  • Microphone array acquisition module with long pickup distance, good noise suppression and better acquisition effect
  • the determination module includes: an input sub-module that can be used for the plurality of voice assistants to input the external instructions; a wake-up determination sub-module that can be used for the plurality of voice assistants to determine whether the external instructions are related to the respective preset first The wake-up marker is consistent; the wake-up sub-module can be used to wake up the corresponding voice assistant when the external instruction is consistent with the first wake-up marker; the interrupt sub-module can be used when the external instruction is consistent with the first When a wake-up marker is inconsistent, the response process of the corresponding voice assistant is interrupted.
  • the device includes:
  • a microphone array acquisition module 700 configured to acquire an externally input instruction through the microphone array
  • An obtaining module 701, configured to obtain external instructions through multiple current voice assistants of the electronic device
  • the determining module 702 is configured to determine, by the plurality of voice assistants, an interrupt response process or wake-up according to the external instruction, wherein the wake-up functions of the plurality of voice assistants are in an enabled state.
  • the determination module 702 includes:
  • a wake-up sub-module 705, configured to wake up a corresponding voice assistant when the external instruction is consistent with the first wake-up marker
  • An interruption submodule 706 is configured to interrupt a response process of a corresponding voice assistant when the external instruction is inconsistent with the first wake-up marker.
  • the second embodiment is a method embodiment corresponding to this embodiment, and this embodiment can be implemented in cooperation with the second embodiment. Relevant technical details mentioned in the second embodiment are still valid in this embodiment, and in order to reduce repetition, details are not repeated here. Accordingly, related technical details mentioned in this embodiment can also be applied in the second embodiment.
  • FIG. 8 is a schematic structural diagram of the device.
  • the device in this embodiment is applied to an electronic device that integrates multiple voice assistants.
  • This embodiment is improved on the basis of the seventh embodiment. The improvements are as follows:
  • the determination module further includes a trigger event determination sub-module, which can be used to determine whether the external instruction includes a voice trigger event, thereby determining whether to trigger the wake-up determination sub-module to work.
  • the device includes:
  • a microphone array acquisition module 800 configured to acquire an externally input instruction through the microphone array
  • An obtaining module 801, configured to obtain external instructions through multiple current voice assistants of the electronic device
  • the determining module 802 is configured to determine, by the plurality of voice assistants, an interrupt response process or wake-up according to the external instruction, wherein the wake-up functions of the plurality of voice assistants are in an enabled state.
  • the determination module 802 includes:
  • a trigger event determination sub-module 804 configured to determine whether the external instruction includes a voice trigger event
  • a wake-up judging sub-module 807 configured to: when the external instruction includes the voice trigger event, the plurality of voice assistants determine whether the external instruction is consistent with a respective preset first wake-up marker;
  • a wake-up submodule 808, configured to wake up a corresponding voice assistant when the external instruction is consistent with the first wake-up marker
  • An interruption sub-module 809 is configured to interrupt a response process of a corresponding voice assistant when the external instruction is inconsistent with the first wake-up marker.
  • the trigger event judgment sub-module 804 includes:
  • a keyword judging sub-module 805, configured to judge whether the external instruction includes a keyword associated with an identifier of any one of the plurality of voice assistants;
  • a trigger event confirmation submodule 806 is configured to determine that the external instruction includes the voice trigger event when the external instruction includes a keyword associated with an identifier of any one of the multiple voice assistants.
  • the third embodiment is a method embodiment corresponding to this embodiment, and this embodiment can be implemented in cooperation with the third embodiment.
  • the related technical details mentioned in the third embodiment are still valid in this embodiment. In order to reduce repetition, details are not repeated here. Accordingly, the related technical details mentioned in this embodiment can also be applied in the third embodiment.
  • FIG. 9 is a schematic structural diagram of the device.
  • the device in this embodiment is applied to an electronic device that integrates multiple voice assistants.
  • This embodiment is improved based on the eighth embodiment. The improvements are as follows:
  • the wake-up determination sub-module includes:
  • An association submodule configured to associate the keyword with a preset corresponding second wake-up marker when it is determined that the external instruction includes the voice trigger event
  • the marker judging sub-module may be used for the plurality of voice assistants to determine whether the second wake-up marker is consistent with a respective preset first wake-up marker.
  • the device includes:
  • a microphone array acquisition module 900 configured to acquire an externally input instruction through the microphone array
  • An obtaining module 901 configured to obtain external instructions through multiple current voice assistants of the electronic device
  • a determining module 902 is configured to determine, by the multiple voice assistants, an interrupt response process or wake-up according to the external instruction, wherein the wake-up functions of the multiple voice assistants are in an enabled state.
  • the determination module 902 includes:
  • a trigger event determination sub-module 904 configured to determine whether the external instruction includes a voice trigger event
  • a wake-up judging sub-module 907 configured to: when the external instruction includes the voice trigger event, the plurality of voice assistants determine whether the external instruction is consistent with a respective preset first wake-up identifier;
  • a wake-up sub-module 910 configured to wake up a corresponding voice assistant when the external instruction is consistent with the first wake-up marker
  • An interruption sub-module 911 is configured to interrupt a response process of a corresponding voice assistant when the external instruction is inconsistent with the first wake-up marker.
  • the trigger event judgment sub-module 904 includes:
  • a keyword judging sub-module 905, configured to judge whether the external instruction includes a keyword associated with an identifier of any one of the plurality of voice assistants;
  • a trigger event confirmation submodule 906 is configured to determine that the external instruction includes the voice trigger event when the external instruction includes a keyword associated with an identifier of any of the multiple voice assistants.
  • the wake-up judgment sub-module 907 includes:
  • An association submodule 908 configured to associate the keyword with a preset corresponding second wake-up marker when it is determined that the external instruction includes the voice trigger event
  • the marker judging sub-module 909 is configured for the plurality of voice assistants to determine whether the second wake-up marker is consistent with a respective preset first wake-up marker.
  • the fourth embodiment is a method embodiment corresponding to this embodiment, and this embodiment can be implemented in cooperation with the fourth embodiment.
  • the relevant technical details mentioned in the fourth embodiment are still valid in this embodiment, and in order to reduce repetition, they are not repeated here. Accordingly, related technical details mentioned in this embodiment can also be applied in the fourth embodiment.
  • the tenth embodiment of the present invention relates to a multi-voice assistant control system
  • FIG. 10 is a schematic structural diagram of the system.
  • the system described in this embodiment includes a cloud server 1000 and a multi-voice assistant control device.
  • the multi-voice assistant control device in this embodiment is improved on the basis of the multi-voice assistant control device described in the sixth embodiment.
  • the main improvements are:
  • the device also includes:
  • the transceiver module may be configured to send an external instruction to the cloud server and / or receive a processed result of the cloud server through a voice assistant that has been determined to be woken up;
  • the response module may be configured to respond to the processed result of the cloud server received by the transceiver module through the voice assistant that has been determined to be woken up.
  • the device includes:
  • An obtaining module 1001 configured to obtain external instructions through multiple current voice assistants of the electronic device
  • a determining module 1002 is configured to perform interrupt response process or wake-up according to the external instruction determined by the multiple voice assistants, wherein the wake-up functions of the multiple voice assistants are in an enabled state;
  • the transceiver module 1003 may be configured to send an external instruction to the cloud server 1000 and / or receive a processed result of the cloud server 1000 through a voice assistant that has been determined to be woken up;
  • the response module 1004 may be configured to respond to the processed result of the cloud server 1000 received by the transceiver module 1003 through the voice assistant that has been determined to wake up.
  • the fifth embodiment is a method embodiment corresponding to this embodiment, and this embodiment can be implemented in cooperation with the fifth embodiment.
  • the related technical details mentioned in the fifth embodiment are still valid in this embodiment, and in order to reduce repetition, details are not repeated here. Accordingly, the related technical details mentioned in this embodiment can also be applied in the fifth embodiment.
  • An embodiment of the present invention also relates to a multi-voice assistant control device, which is applied to an electronic device integrating multiple voice assistants, and includes:
  • a memory for storing the processor-executable instructions
  • the processor is configured to:
  • the plurality of voice assistants determine whether to perform an interrupt response process or wake up according to the external instruction, and the wake-up functions of the plurality of voice assistants are all in an enabled state.
  • An embodiment of the present invention also relates to a computer-readable storage medium having computer program instructions stored thereon, which when executed by a processor, implement the steps of the method described in the first embodiment.
  • the method embodiments of the present invention can be implemented in software, hardware, firmware, and the like. Regardless of whether the present invention is implemented in software, hardware, or firmware, the instruction code can be stored in any type of computer-accessible memory (for example, permanent or modifiable, volatile or nonvolatile, solid state Or non-solid, fixed or replaceable media, etc.).
  • the memory may be, for example, Programmable Array Logic (PAL), Random Access Memory (RAM), Programmable Read Only Memory (PROM) "), Read-Only Memory (“ ROM “), Electrically Erasable Programmable ROM (“EEPROM “), magnetic disks, optical disks, digital Versatile Discs , “DVD” for short) and so on.
  • PAL Programmable Array Logic
  • RAM Random Access Memory
  • PROM Programmable Read Only Memory
  • ROM Read-Only Memory
  • EEPROM Electrically Erasable Programmable ROM
  • each module mentioned in the embodiments of each device of the present invention is a logic module.
  • a logic module may be a physical module, or a part of a physical module, and may also be a plurality of physical modules.
  • the combined implementation of modules, the physical implementation of these logical modules themselves is not the most important, and the combination of functions implemented by these logical modules is the key to solving the technical problem proposed by the present invention.
  • the above-mentioned device embodiments of the present invention do not introduce modules that are not closely related to solving the technical problems proposed by the present invention, which does not indicate that there are no other embodiments of the above-mentioned device. Module.

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种多语音助手控制方法、装置、系统及计算机可读存储介质。多语音助手控制方法应用于集成多个语音助手的电子设备，方法步骤包括：通过电子设备当前的多个语音助手获取外部指令（100）；多个语音助手根据外部指令判定进行中断响应进程或唤醒（101），其中，多个语音助手的唤醒功能均处于使能状态。该方法可以在多个语音助手之间随时进行语音切换，不会出现多个语音助手同时处于唤醒工作状态，避免了互相干扰，影响用户体验。

Description

一种多语音助手控制方法、装置、系统及计算机可读存储介质 技术领域
本发明涉及语音技术领域，特别涉及一种多语音助手控制方法、装置、系统及计算机可读存储介质。
背景技术
随着语音识别技术的发展和人们对于语音交互的需求增加，语音助手这一智能型应用应运而生。在相关技术中，语音助手可以与用户实现智能对话或即时问答，以帮助用户解决疑难问题。然而，现有电子设备通常只内置有一个语音助手，即便内置有多个语音助手，有的会出现在一语音助手工作时其他语音助手被唤醒工作，多个语音助手同时工作导致冲突(互相干扰)，用户体验不理想。
发明内容
本发明的目的在于提供一种多语音助手控制方法、装置、系统及计算机可读存储介质，用以解决现有电子设备多个语音助手同时工作导致冲突(互相干扰)，用户体验不理想的问题。
在本发明中，第一方面提供了一种多语音助手控制方法，所述方法应用于集成多个语音助手的电子设备，所述方法步骤包括：
通过所述电子设备当前的多个语音助手获取外部指令；
所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
较佳地，所述步骤“所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒”包括步骤：
所述多个语音助手输入所述外部指令；
所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致；
如果所述外部指令与所述第一唤醒标记符一致，则唤醒相应的语音助手， 否则，中断相应语音助手的响应进程。
较佳地，所述步骤“所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致”之前包括步骤：
判断所述外部指令是否包含语音触发事件；
如果所述外部指令包含所述语音触发事件，则执行所述步骤“所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致”。
较佳地，所述步骤“判断所述外部指令是否包含语音触发事件”包括步骤：
判断所述外部指令是否包括与任一所述多个语音助手的标识相关联的关键词；
如果所述外部指令中包括与任一所述多个语音助手的标识相关联的关键词，则确定所述外部指令包含所述语音触发事件。
较佳地，所述步骤“所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致”包括步骤：
当确定所述外部指令包含所述语音触发事件时，将所述关键词与预设的对应的第二唤醒标记符相关联；
所述多个语音助手判断所述第二唤醒标记符是否与各自预设的第一唤醒标记符一致；
如果所述第二唤醒标记符与各自预设的第一唤醒标记符一致，则执行所述步骤“唤醒相应的语音助手”，否则，执行所述步骤“中断相应语音助手的响应进程”。
较佳地，所述步骤“所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒”包括步骤：
一个语音助手根据所述外部指令判定进行唤醒，同时其他语音助手根据所述外部指令判定进行中断响应进程。
较佳地，所述步骤“通过所述电子设备当前的多个语音助手获取外部指令”之前包括步骤：
通过麦克风阵列采集外部输入的指令。
较佳地，所述步骤“所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒”之后包括步骤：
已判定唤醒的语音助手将所述外部指令通过网络发送给云端服务器处理；
通过所述已判定唤醒的语音助手将所述云端服务器处理后的结果进行响应。
本发明第二方面提供了一种多语音助手控制装置，所述装置应用于集成多个语音助手的电子设备，所述装置包括：
获取模块，用于通过所述电子设备当前的多个语音助手获取外部指令；
判定模块，用于所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
较佳地，所述判定模块包括：
输入子模块，用于所述多个语音助手输入所述外部指令；
唤醒判断子模块，用于所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致；
唤醒子模块，用于当所述外部指令与所述第一唤醒标记符一致时，唤醒相应的语音助手；
中断子模块，用于当所述外部指令与所述第一唤醒标记符不一致时，中断相应语音助手的响应进程。
较佳地，所述判定模块包括：
触发事件判断子模块，用于判断所述外部指令是否包含语音触发事件，
所述唤醒判断子模块，用于当所述外部指令包含所述语音触发事件时，所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致。
较佳地，所述触发事件判断子模块包括：
关键词判断子模块，用于判断所述外部指令是否包括与任一所述多个语音助手的标识相关联的关键词；
触发事件确认子模块，用于当所述外部指令中包括与任一所述多个语音助手的标识相关联的关键词时，确定所述外部指令包含所述语音触发事件。
较佳地，所述唤醒判断子模块包括：
关联子模块，用于当确定所述外部指令包含所述语音触发事件时，将所述关键词与预设的对应的第二唤醒标记符相关联；
标记符判断子模块，用于所述多个语音助手判断所述第二唤醒标记符是否与各自预设的第一唤醒标记符一致；
所述判定模块包括：
所述唤醒子模块，用于当所述第二唤醒标记符与各自预设的第一唤醒标记符一致时，唤醒相应的语音助手；
所述中断子模块，用于当所述第二唤醒标记符与各自预设的第一唤醒标记符不一致时，中断相应语音助手的响应进程。
较佳地，所述装置包括：
麦克风阵列采集模块，用于通过麦克风阵列采集外部输入的指令。
本发明第三方面提供了一种多语音助手控制系统，所述系统包括：
云端服务器和第二方面所述的多语音助手控制装置；
所述多语音助手控制装置包括：
收发模块，用于通过已判定唤醒的语音助手向所述云端服务器发送外部指令和/或接收所述云端服务器处理后的结果；
响应模块，用于通过所述已判定唤醒的语音助手对所述收发模块所接收的云端服务器处理后的结果进行响应。
本发明还提供了一种多语音助手控制装置，所述装置应用于集成多个语音助手的电子设备，包括：
处理器；
用于存储所述处理器可执行指令的存储器；
其中，所述处理器被配置为：
通过所述电子设备当前的多个语音助手获取外部指令；
所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
本发明还提供了一种计算机可读存储介质，其上存储有计算机程序指令，该程序指令被处理器执行时实现第一方面所述方法的步骤。
本发明实施方式与现有技术相比，主要区别及其效果在于：
本发明一语音助手根据指令进行唤醒,同时其他语音助手根据指令中断响应进程,不会出现多个语音助手同时处于唤醒工作状态，避免了互相干扰，影响用户体验。
进一步地，预设唤醒词可以由用户自定义，不但丰富了语音助手使用方法，避免了语音助手角色的单一。
进一步地，通过语音指令变换来切换唤醒助手，还解决了现有电子设备切 换语音助手时需要手动使能语音助手，切换方式不够灵活的技术问题。
进一步地，用户可以根据自身喜好和接受度来从多个语音助手中选择一语音助手进行问答，可以适应不同用户使用，人机交互更有趣味性。
进一步地，麦克风阵列采集外部输入的指令，拾音距离远、噪声抑制佳、采集效果更好。
应理解，在本发明范围内中，本发明的上述各技术特征和在下文(如实施例)中具体描述的各技术特征之间都可以互相组合，从而构成新的或优选的技术方案。限于篇幅，在此不再一一赘述。
附图说明
图1是本发明第一实施方式中多语音助手控制方法的流程示意图。
图2是本发明第二实施方式中多语音助手控制方法的流程示意图。
图3是本发明第三实施方式中多语音助手控制方法的流程示意图。
图4是本发明第四实施方式中多语音助手控制方法的流程示意图。
图5是本发明第五实施方式中多语音助手控制方法的流程示意图。
图6是本发明第六实施方式中多语音助手控制装置的结构示意图。
图7是本发明第七实施方式中多语音助手控制装置的结构示意图。
图8是本发明第八实施方式中多语音助手控制装置的结构示意图。
图9是本发明第九实施方式中多语音助手控制装置的结构示意图。
图10是本发明第十实施方式中多语音助手控制系统的结构示意图。
具体实施方式
在以下的叙述中，为了使读者更好地理解本申请而提出了许多技术细节。但是，本领域的普通技术人员可以理解，即使没有这些技术细节和基于以下各实施方式的种种变化和修改，也可以实现本申请所要求保护的技术方案。
为使本发明的目的、技术方案和优点更加清楚，下面将结合附图对本发明的实施方式作进一步地详细描述。
术语解释
响应进程：本文所提及的“响应进程”是指：语音助手被唤醒后，根据用户的指令进行回答(响应)的音频进程，本文中，响应进程与音频进程可互换使用。
监听状态：本文所提及的“监听状态”是指：语音助手在被唤醒之前，等待唤醒词激活的状态，处于监听状态的语音助手无响应进程。
工作状态：本文所提及的“工作状态”是指：语音助手的音频进程处于活跃状态。
重新唤醒：本文所提及的“重新唤醒”是指：语音助手的音频进程处于活跃状态时，被唤醒词重新激活，之前处于活跃状态的音频进程被中断。
本发明第一实施方式涉及一种多语音助手控制方法，图1是该方法的流程图。
如图所示，本实施方式中方法应用于集成多个语音助手的电子设备。
在步骤100中，通过所述电子设备当前的多个语音助手获取外部指令。
此后进入步骤101：所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
优选地，步骤101包括步骤：
一个语音助手根据所述外部指令判定进行唤醒，同时其他语音助手根据所述外部指令判定进行中断响应进程。
本实施方式中，电子设备当前的多个语音助手获取外部指令，该外部指令可以是语音信息。多个语音助手根据外部指令判定进行中断响应进程或唤醒，例如：一语音助手处于监听状态(未被唤醒)根据外部指令进行唤醒，同时一语音助手处于工作状态根据外部指令进行中断响应进程；又例如：一语音助手处于工作状态根据外部指令进行唤醒(重新唤醒)，同时一语音助手处于监听状态(未被唤醒)根据外部指令进行中断响应进程(保持监听状态)，保证在同一时刻不会出现多个语音助手同时工作导致冲突(互相干扰)的情况。
本发明第二实施方式涉及一种多语音助手控制方法，图2是该方法的流程图。
如图所示，本实施方式中方法应用于集成多个语音助手的电子设备。
在步骤200中，通过麦克风阵列采集外部输入的指令。
此后进入步骤201：通过所述电子设备当前的多个语音助手获取外部指令。
此后进入步骤202：所述多个语音助手输入所述外部指令。
此后进入步骤203：所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致，如果是，则执行步骤204：唤醒相应的语音助手，否则执行步骤205：中断相应语音助手的响应进程。
其中，步骤201可以实现步骤100的功能，步骤202、步骤203、步骤204和步骤205可以实现步骤101的功能。
本实施方式中，麦克风阵列可以采用多种形式：线形、环形和球形，例如：2麦克风阵列、6+1麦克风阵列和8+1麦克风阵列，拾音距离远、噪声抑制佳、采集效果更好。
本发明第三实施方式涉及一种多语音助手控制方法，图3是该方法的流程图。
如图所示，本实施方式中方法应用于集成多个语音助手的电子设备。
在步骤300中，通过麦克风阵列采集外部输入的指令。
此后进入步骤301：通过所述电子设备当前的多个语音助手获取外部指令。
此后进入步骤302：所述多个语音助手输入所述外部指令。
此后进入步骤303：判断所述外部指令是否包括与任一所述多个语音助手的标识相关联的关键词，如果是，则执行步骤304：所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致，否则执行上述步骤300。
步骤304：所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致，如果是，则执行步骤305：唤醒相应的语音助手，否则执行步骤306：中断相应语音助手的响应进程。
其中，步骤300可以实现步骤200的功能，步骤301可以实现步骤201的功能，步骤302可以实现步骤202的功能，步骤304可以实现步骤203的功能，步骤305可以实现步骤204的功能，步骤306可以实现步骤205的功能。
本发明第四实施方式涉及一种多语音助手控制方法，图4是该方法的流程图。
如图所示，本实施方式中方法应用于集成多个语音助手的电子设备。
在步骤400中，通过麦克风阵列采集外部输入的指令。
此后进入步骤401：通过所述电子设备当前的多个语音助手获取外部指令。
此后进入步骤402：所述多个语音助手输入所述外部指令。
此后进入步骤403：判断所述外部指令是否包括与任一所述多个语音助手的标识相关联的关键词，如果是，则执行步骤404：将所述关键词与预设的对应的第二唤醒标记符相关联，否则执行上述步骤400。
步骤404之后进入步骤405：所述多个语音助手判断所述第二唤醒标记符是否与各自预设的第一唤醒标记符一致，如果是，则执行步骤406：唤醒相应的语音助手，否则执行步骤407：中断相应语音助手的响应进程。
其中，步骤400可以实现步骤300的功能，步骤401可以实现步骤301的功能，步骤402可以实现步骤302的功能，步骤403可以实现步骤303的功能，步骤404、步骤405可以实现步骤304的功能，步骤406可以实现步骤305的功能，步骤407可以实现步骤306的功能。
本实施方式中，多个语音助手均设有与之对应的唤醒词(关键词)，语音助手与预设的唤醒词(关键词)之间存在预设对应关系，例如：语音助手A的唤醒词是小A，语音助手B的唤醒词是小B，语音助手C的唤醒词是小C，当多个语音助手收到的外部指令包含某个语音助手的唤醒词时，则该语音助手被唤醒，其他语音助手中断相应语音助手的响应进程。
本实施方式对应一种情景如下：
麦克风阵列采集外部输入的指令包含“小A”，多个语音助手收到该指令并判定该指令中包含唤醒词(关键词)：“小A”，将该唤醒词与预设的对应的第二唤醒标记符相关联(小A对应的第二唤醒标记符为00，小B对应的第二唤醒标记符为01，小C对应的第二唤醒标记符为02，)，将关联后的第二唤醒标记符(00)与各自预设的第一唤醒标记符进行比对(语音助手A预设的第一唤醒标记符为00，语音助手B预设的第一唤醒标记符为01，语音助手C预设的第一唤醒标记符为02)，语音助手A判定第二唤醒标记符(00)与自身预设的第一唤醒标记符(00)一致，则进行唤醒(如果语音助手A之前是监听状态，则被唤醒工作，如果之前是工作状态，则被重新唤醒)，语音助手B和语音助手C判定第二唤醒标记符(00)与自身预设的第一唤醒标记符(01)、(02)不一致，则中断相应语音助手的响应进程(如果语音助手B/语音助手C之前是监听状态，则保持监听状态，如果之前是工作状态，则中断相应语音助手的响应进程)。
例如：百度语音助手的唤醒词(关键词)是“小度”，亚马逊语音助手的唤醒词(关键词)是“Alexa”，将百度语音助手和亚马逊语音助手集成并应用于电子设备。当用户对该电子设备首先发出“小度”指令时，百度语音助手被唤醒，亚马逊语音助手保持监听状态；之后用户对该电子设备发出“Alexa”指令时，百度语音助手中断响应进程，亚马逊语音助手被唤醒。也可以是当用户对该电子设备首先发出“小度”指令时，百度语音助手被唤醒，亚马逊语音助手保持监听状态；之后用户对该电子设备再次发出“小度”指令时，百度语音助手重新被唤醒，亚马逊语音助手保持监听状态。
需要说明的是，语音助手的唤醒词(关键词)可以是各个语音助手的名称，还可以是用户根据实际需求自定义的词句。若预设唤醒词和语音助手之间不存在预设对应关系，则用户自定义的预设唤醒词不具体指定唤醒哪一语音助手，系统可以直接唤醒默认语音助手或者提示用户再次选择具体唤醒哪一语音助手。
采用上述方法，可以在多个语音助手之间随时进行语音切换，不会出现多个语音助手同时处于唤醒工作状态，避免了互相干扰，影响用户体验。同时，预设唤醒词可以由用户自定义，不但丰富了语音助手使用方法，避免了语音助手角色的单一，通过语音指令变换来切换唤醒助手，还解决了现有电子设备切换语音助手时需要手动使能语音助手，切换方式不够灵活的技术问题。
本发明第五实施方式涉及一种多语音助手控制方法，图5是该方法的流程图。
如图所示，本实施方式中方法应用于集成多个语音助手的电子设备。
在步骤500中，通过所述电子设备当前的多个语音助手获取外部指令。
此后进入步骤501：所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒。
此后进入步骤502：已判定唤醒的语音助手将所述外部指令通过网络发送给云端服务器处理。
此后进入步骤503：通过所述已判定唤醒的语音助手将所述云端服务器处理后的结果进行响应。
其中，步骤500可以实现步骤100的功能，步骤501可以实现步骤101的功能。
本实施方式对应一种情景如下：
将百度语音助手和亚马逊语音助手集成并应用于电子设备(百度语音助手的唤醒词(关键词)是“小度”，亚马逊语音助手的唤醒词(关键词)是“Alexa”)。
用户首先对电子设备发出“小度，今天上海天气”指令，百度语音助手被唤醒(亚马逊语音助手保持监听)，百度语音助手将该指令通过网络上传至云端服务器，云端服务器根据该指令进行处理并将结果(可以是语音包)发回给百度语音助手，百度语音助手将该处理的结果进行响应(发出“今天上海天气多云，25°”)。之后用户对电子设备发出“Alexa，What’s the weather of ShangHai today”指令，亚马逊语音助手被唤醒(百度语音助手中断之前的音频/响应进程)，亚马逊语音助手将该指令通过网络上传至云端服务器，云端服务器根据该指令进行处理并将结果(可以是语音包)发回给亚马逊语音助手，亚马逊语音助手将该处理的结果进行响应(发出“Today t he weather of ShangHai is cloudy”)。
采用上述方法，用户可以根据自身喜好和接受度来从多个语音助手中选择一语音助手进行问答(QA)，可以适应不同用户使用。
本发明第六实施方式涉及一种多语音助手控制装置，图6是该装置的结构示意图。
如图所示，本实施方式中装置应用于集成多个语音助手的电子设备。
所述装置包括：
获取模块600，用于通过所述电子设备当前的多个语音助手获取外部指令；
判定模块601，用于所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
第一实施方式是与本实施方式相对应的方法实施方式，本实施方式可与第一实施方式互相配合实施。第一实施方式中提到的相关技术细节在本实施方式中依然有效，为了减少重复，这里不再赘述。相应地，本实施方式中提到的相关技术细节也可应用在第一实施方式中。
本发明第七实施方式涉及一种多语音助手控制装置，图7是该装置的结构示意图。
如图所示，本实施方式中装置应用于集成多个语音助手的电子设备，本实施方式在第六实施方式基础上进行了改进，改进之处在于：
本实施方式中，装置还包括：
麦克风阵列采集模块，拾音距离远、噪声抑制佳、采集效果更好；
判定模块包括：输入子模块，可以用于所述多个语音助手输入所述外部指令；唤醒判断子模块，可以用于所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致；唤醒子模块，可以用于当所述外部指令与所述第一唤醒标记符一致时，唤醒相应的语音助手；中断子模块，可以用于当所述外部指令与所述第一唤醒标记符不一致时，中断相应语音助手的响应进程。
具体地说：
所述装置包括：
麦克风阵列采集模块700，用于通过麦克风阵列采集外部输入的指令；
获取模块701，用于通过所述电子设备当前的多个语音助手获取外部指令；
判定模块702，用于所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
其中，判定模块702包括：
输入子模块703，用于所述多个语音助手输入所述外部指令；
唤醒判断子模块704，用于所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致；
唤醒子模块705，用于当所述外部指令与所述第一唤醒标记符一致时，唤醒相应的语音助手；
中断子模块706，用于当所述外部指令与所述第一唤醒标记符不一致时，中断相应语音助手的响应进程。
第二实施方式是与本实施方式相对应的方法实施方式，本实施方式可与第二实施方式互相配合实施。第二实施方式中提到的相关技术细节在本实施方式中依然有效，为了减少重复，这里不再赘述。相应地，本实施方式中提到的相关技术细节也可应用在第二实施方式中。
本发明第八实施方式涉及一种多语音助手控制装置，图8是该装置的结构示意图。
如图所示，本实施方式中装置应用于集成多个语音助手的电子设备，本实施方式在第七实施方式基础上进行了改进，改进之处在于：
本实施方式中，判定模块还包括：触发事件判断子模块，可以用于判断所述外部指令是否包含语音触发事件，从而确定是否触发唤醒判断子模块工作。
具体地说：
所述装置包括：
麦克风阵列采集模块800，用于通过麦克风阵列采集外部输入的指令；
获取模块801，用于通过所述电子设备当前的多个语音助手获取外部指令；
判定模块802，用于所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
其中，判定模块802包括：
输入子模块803，用于所述多个语音助手输入所述外部指令；
触发事件判断子模块804，用于判断所述外部指令是否包含语音触发事件；
唤醒判断子模块807，用于当所述外部指令包含所述语音触发事件时，所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致；
唤醒子模块808，用于当所述外部指令与所述第一唤醒标记符一致时，唤醒相应的语音助手；
中断子模块809，用于当所述外部指令与所述第一唤醒标记符不一致时，中断相应语音助手的响应进程。
其中，触发事件判断子模块804包括：
关键词判断子模块805，用于判断所述外部指令是否包括与任一所述多个语音助手的标识相关联的关键词；
触发事件确认子模块806，用于当所述外部指令中包括与任一所述多个语音助手的标识相关联的关键词时，确定所述外部指令包含所述语音触发事件。
第三实施方式是与本实施方式相对应的方法实施方式，本实施方式可与第三实施方式互相配合实施。第三实施方式中提到的相关技术细节在本实施方式中依然有效，为了减少重复，这里不再赘述。相应地，本实施方式中提到的相关技术细节也可应用在第三实施方式中。
本发明第九实施方式涉及一种多语音助手控制装置，图9是该装置的结构示意图。
如图所示，本实施方式中装置应用于集成多个语音助手的电子设备，本实施方式在第八实施方式基础上进行了改进，改进之处在于：
本实施方式中，唤醒判断子模块包括：
关联子模块，可以用于当确定所述外部指令包含所述语音触发事件时，将 所述关键词与预设的对应的第二唤醒标记符相关联；
标记符判断子模块，可以用于所述多个语音助手判断所述第二唤醒标记符是否与各自预设的第一唤醒标记符一致。
具体地说：
所述装置包括：
麦克风阵列采集模块900，用于通过麦克风阵列采集外部输入的指令；
获取模块901，用于通过所述电子设备当前的多个语音助手获取外部指令；
判定模块902，用于所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
其中，判定模块902包括：
输入子模块903，用于所述多个语音助手输入所述外部指令；
触发事件判断子模块904，用于判断所述外部指令是否包含语音触发事件；
唤醒判断子模块907，用于当所述外部指令包含所述语音触发事件时，所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致；
唤醒子模块910，用于当所述外部指令与所述第一唤醒标记符一致时，唤醒相应的语音助手；
中断子模块911，用于当所述外部指令与所述第一唤醒标记符不一致时，中断相应语音助手的响应进程。
其中，触发事件判断子模块904包括：
关键词判断子模块905，用于判断所述外部指令是否包括与任一所述多个语音助手的标识相关联的关键词；
触发事件确认子模块906，用于当所述外部指令中包括与任一所述多个语音助手的标识相关联的关键词时，确定所述外部指令包含所述语音触发事件。
唤醒判断子模块907包括：
关联子模块908，用于当确定所述外部指令包含所述语音触发事件时，将所述关键词与预设的对应的第二唤醒标记符相关联；
标记符判断子模块909，用于所述多个语音助手判断所述第二唤醒标记符是否与各自预设的第一唤醒标记符一致。
第四实施方式是与本实施方式相对应的方法实施方式，本实施方式可与第四实施方式互相配合实施。第四实施方式中提到的相关技术细节在本实施方式 中依然有效，为了减少重复，这里不再赘述。相应地，本实施方式中提到的相关技术细节也可应用在第四实施方式中。
本发明第十实施方式涉及一种多语音助手控制系统，图10是该系统的结构示意图。
如图所示，本实施方式中所述系统包括：云端服务器1000和多语音助手控制装置。
本实施方式中的多语音助手控制装置是在第六实施方式中所述的多语音助手控制装置基础上进行改进的，主要改进之处在于：
装置还包括：
收发模块，可以用于通过已判定唤醒的语音助手向所述云端服务器发送外部指令和/或接收所述云端服务器处理后的结果；
响应模块，可以用于通过所述已判定唤醒的语音助手对所述收发模块所接收的云端服务器处理后的结果进行响应。
具体地说：
本实施方式中，装置包括：
获取模块1001，用于通过所述电子设备当前的多个语音助手获取外部指令；
判定模块1002，用于所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态；
收发模块1003，可以用于通过已判定唤醒的语音助手向所述云端服务器1000发送外部指令和/或接收所述云端服务器1000处理后的结果；
响应模块1004，可以用于通过所述已判定唤醒的语音助手对所述收发模块1003所接收的云端服务器1000处理后的结果进行响应。
第五实施方式是与本实施方式相对应的方法实施方式，本实施方式可与第五实施方式互相配合实施。第五实施方式中提到的相关技术细节在本实施方式中依然有效，为了减少重复，这里不再赘述。相应地，本实施方式中提到的相关技术细节也可应用在第五实施方式中。
本发明一实施方式还涉及一种多语音助手控制装置，所述装置应用于集成多个语音助手的电子设备，包括：
处理器；
用于存储所述处理器可执行指令的存储器；
其中，所述处理器被配置为：
通过所述电子设备当前的多个语音助手获取外部指令；
所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
本发明一实施方式还涉及一种计算机可读存储介质，其上存储有计算机程序指令，该程序指令被处理器执行时实现第一实施方式中所述方法的步骤。
本发明的各方法实施方式均可以以软件、硬件、固件等方式实现。不管本发明是以软件、硬件、还是固件方式实现，指令代码都可以存储在任何类型的计算机可访问的存储器中(例如永久的或者可修改的，易失性的或者非易失性的，固态的或者非固态的，固定的或者可更换的介质等等)。同样，存储器可以例如是可编程阵列逻辑(Programmable Array Logic，简称“PAL”)、随机存取存储器(Random Access Memory，简称“RAM”)、可编程只读存储器(Programmable Read Only Memory，简称“PROM”)、只读存储器(Read-Only Memory，简称“ROM”)、电可擦除可编程只读存储器(Electrically Erasable Programmable ROM，简称“EEPROM”)、磁盘、光盘、数字通用光盘(Digital Versatile Disc，简称“DVD”)等等。
需要说明的是，本发明各设备实施方式中提到的各模块都是逻辑模块，在物理上，一个逻辑模块可以是一个物理模块，也可以是一个物理模块的一部分，还可以以多个物理模块的组合实现，这些逻辑模块本身的物理实现方式并不是最重要的，这些逻辑模块所实现的功能的组合才是解决本发明所提出的技术问题的关键。此外，为了突出本发明的创新部分，本发明上述各设备实施方式并没有将与解决本发明所提出的技术问题关系不太密切的模块引入，这并不表明上述设备实施方式并不存在其它的模块。
需要说明的是，在本专利的说明书中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有 的要素。在没有更多限制的情况下，由语句“包括一个”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
虽然通过参照本发明的某些优选实施方式，已经对本发明进行了图示和描述，但本领域的普通技术人员应该明白，可以在形式上和细节上对其作各种改变，而不偏离本发明的精神和范围。

Claims (17)

  1. 一种多语音助手控制方法，其特征在于，所述方法应用于集成多个语音助手的电子设备，所述方法步骤包括：
    通过所述电子设备当前的多个语音助手获取外部指令；
    所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
  2. 根据权利要求1所述的方法，其特征在于，所述步骤“所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒”包括步骤：
    所述多个语音助手输入所述外部指令；
    所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致；
    如果所述外部指令与所述第一唤醒标记符一致，则唤醒相应的语音助手，否则，中断相应语音助手的响应进程。
  3. 根据权利要求2所述的方法，其特征在于，所述步骤“所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致”之前包括步骤：
    判断所述外部指令是否包含语音触发事件；
    如果所述外部指令包含所述语音触发事件，则执行所述步骤“所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致”。
  4. 根据权利要求3所述的方法，其特征在于，所述步骤“判断所述外部指令是否包含语音触发事件”包括步骤：
    判断所述外部指令是否包括与任一所述多个语音助手的标识相关联的关键词；
    如果所述外部指令中包括与任一所述多个语音助手的标识相关联的关键词，则确定所述外部指令包含所述语音触发事件。
  5. 根据权利要求4所述的方法，其特征在于，所述步骤“所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致”包括步骤：
    当确定所述外部指令包含所述语音触发事件时，将所述关键词与预设的对应的第二唤醒标记符相关联；
    所述多个语音助手判断所述第二唤醒标记符是否与各自预设的第一唤醒标记符一致；
    如果所述第二唤醒标记符与各自预设的第一唤醒标记符一致，则执行所述步骤“唤醒相应的语音助手”，否则，执行所述步骤“中断相应语音助手的响应进程”。
  6. 根据权利要求1所述的方法，其特征在于，所述步骤“所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒”包括步骤：
    一个语音助手根据所述外部指令判定进行唤醒，同时其他语音助手根据所述外部指令判定进行中断响应进程。
  7. 根据权利要求1所述的方法，其特征在于，所述步骤“通过所述电子设备当前的多个语音助手获取外部指令”之前包括步骤：
    通过麦克风阵列采集外部输入的指令。
  8. 根据权利要求1所述的方法，其特征在于，所述步骤“所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒”之后包括步骤：
    已判定唤醒的语音助手将所述外部指令通过网络发送给云端服务器处理；
    通过所述已判定唤醒的语音助手将所述云端服务器处理后的结果进行响应。
  9. 一种多语音助手控制装置，其特征在于，所述装置应用于集成多个语音助手的电子设备，所述装置包括：
    获取模块，用于通过所述电子设备当前的多个语音助手获取外部指令；
    判定模块，用于所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
  10. 根据权利要求9所述的装置，其特征在于，所述判定模块包括：
    输入子模块，用于所述多个语音助手输入所述外部指令；
    唤醒判断子模块，用于所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致；
    唤醒子模块，用于当所述外部指令与所述第一唤醒标记符一致时，唤醒相应的语音助手；
    中断子模块，用于当所述外部指令与所述第一唤醒标记符不一致时，中断相 应语音助手的响应进程。
  11. 根据权利要求10所述的装置，其特征在于，所述判定模块包括：
    触发事件判断子模块，用于判断所述外部指令是否包含语音触发事件，
    所述唤醒判断子模块，用于当所述外部指令包含所述语音触发事件时，所述多个语音助手判断所述外部指令是否与各自预设的第一唤醒标记符一致。
  12. 根据权利要求11所述的装置，其特征在于，所述触发事件判断子模块包括：
    关键词判断子模块，用于判断所述外部指令是否包括与任一所述多个语音助手的标识相关联的关键词；
    触发事件确认子模块，用于当所述外部指令中包括与任一所述多个语音助手的标识相关联的关键词时，确定所述外部指令包含所述语音触发事件。
  13. 根据权利要求12所述的装置，其特征在于，所述唤醒判断子模块包括：
    关联子模块，用于当确定所述外部指令包含所述语音触发事件时，将所述关键词与预设的对应的第二唤醒标记符相关联；
    标记符判断子模块，用于所述多个语音助手判断所述第二唤醒标记符是否与各自预设的第一唤醒标记符一致；
    所述判定模块包括：
    所述唤醒子模块，用于当所述第二唤醒标记符与各自预设的第一唤醒标记符一致时，唤醒相应的语音助手；
    所述中断子模块，用于当所述第二唤醒标记符与各自预设的第一唤醒标记符不一致时，中断相应语音助手的响应进程。
  14. 根据权利要求9所述的装置，其特征在于，所述装置包括：
    麦克风阵列采集模块，用于通过麦克风阵列采集外部输入的指令。
  15. 一种多语音助手控制系统，其特征在于，所述系统包括：
    云端服务器和权利要求9所述的多语音助手控制装置；
    所述多语音助手控制装置包括：
    收发模块，用于通过已判定唤醒的语音助手向所述云端服务器发送外部指令和/或接收所述云端服务器处理后的结果；
    响应模块，用于通过所述已判定唤醒的语音助手对所述收发模块所接收的云端服务器处理后的结果进行响应。
  16. 一种多语音助手控制装置，其特征在于，所述装置应用于集成多个语音助手的电子设备，包括：
    处理器；
    用于存储所述处理器可执行指令的存储器；
    其中，所述处理器被配置为：
    通过所述电子设备当前的多个语音助手获取外部指令；
    所述多个语音助手根据所述外部指令判定进行中断响应进程或唤醒，其中，所述多个语音助手的唤醒功能均处于使能状态。
  17. 一种计算机可读存储介质，其上存储有计算机程序指令，其特征在于，该程序指令被处理器执行时实现权利要求1中所述方法的步骤。
PCT/CN2018/089166 2018-05-31 2018-05-31 一种多语音助手控制方法、装置、系统及计算机可读存储介质 WO2019227370A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2018/089166 WO2019227370A1 (zh) 2018-05-31 2018-05-31 一种多语音助手控制方法、装置、系统及计算机可读存储介质
CN201880091294.XA CN111954868A (zh) 2018-05-31 2018-05-31 一种多语音助手控制方法、装置、系统及计算机可读存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/089166 WO2019227370A1 (zh) 2018-05-31 2018-05-31 一种多语音助手控制方法、装置、系统及计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2019227370A1 true WO2019227370A1 (zh) 2019-12-05

Family

ID=68696754

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/089166 WO2019227370A1 (zh) 2018-05-31 2018-05-31 一种多语音助手控制方法、装置、系统及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN111954868A (zh)
WO (1) WO2019227370A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111462747A (zh) * 2020-04-22 2020-07-28 战炜 一种听力辅助设备及其设定方法
CN114697713A (zh) * 2020-12-29 2022-07-01 深圳Tcl新技术有限公司 语音助手控制方法、装置、存储介质及智能电视
US11700139B2 (en) 2020-11-13 2023-07-11 Haier Us Appliance Solutions, Inc. Virtual microphone input for multiple voice assistants

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112817556A (zh) * 2021-02-04 2021-05-18 青岛海信传媒网络技术有限公司 显示设备上语音方案的切换方法、显示设备及控制装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105975063A (zh) * 2016-04-27 2016-09-28 吴波 一种控制智能终端的方法和装置
CN106453859A (zh) * 2016-09-23 2017-02-22 维沃移动通信有限公司 一种语音控制方法及移动终端
CN107919123A (zh) * 2017-12-07 2018-04-17 北京小米移动软件有限公司 多语音助手控制方法、装置及计算机可读存储介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104345877B (zh) * 2013-08-08 2018-04-27 联想(北京)有限公司 一种信息处理的方法及电子设备
US9318107B1 (en) * 2014-10-09 2016-04-19 Google Inc. Hotword detection on multiple devices
CN106168895A (zh) * 2016-07-07 2016-11-30 北京行云时空科技有限公司 用于智能终端的语音控制方法及智能终端
CN107393548B (zh) * 2017-07-05 2021-05-07 海信视像科技股份有限公司 多个语音助手设备采集的语音信息的处理方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105975063A (zh) * 2016-04-27 2016-09-28 吴波 一种控制智能终端的方法和装置
CN106453859A (zh) * 2016-09-23 2017-02-22 维沃移动通信有限公司 一种语音控制方法及移动终端
CN107919123A (zh) * 2017-12-07 2018-04-17 北京小米移动软件有限公司 多语音助手控制方法、装置及计算机可读存储介质

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111462747A (zh) * 2020-04-22 2020-07-28 战炜 一种听力辅助设备及其设定方法
CN111462747B (zh) * 2020-04-22 2024-04-05 战炜 一种听力辅助设备及其设定方法
US11700139B2 (en) 2020-11-13 2023-07-11 Haier Us Appliance Solutions, Inc. Virtual microphone input for multiple voice assistants
CN114697713A (zh) * 2020-12-29 2022-07-01 深圳Tcl新技术有限公司 语音助手控制方法、装置、存储介质及智能电视
CN114697713B (zh) * 2020-12-29 2024-02-06 深圳Tcl新技术有限公司 语音助手控制方法、装置、存储介质及智能电视

Also Published As

Publication number Publication date
CN111954868A (zh) 2020-11-17

Similar Documents

Publication Publication Date Title
WO2019227370A1 (zh) 一种多语音助手控制方法、装置、系统及计算机可读存储介质
JP7198861B2 (ja) ホームオートメーションのためのインテリジェントアシスタント
TWI665584B (zh) 語音控制系統及方法
CN109410952B (zh) 一种语音唤醒方法、装置及系统
US20200075018A1 (en) Control method of multi voice assistants
WO2018188587A1 (zh) 一种语音响应方法、装置及智能设备
CN106463112B (zh) 语音识别方法、语音唤醒装置、语音识别装置及终端
WO2019007245A1 (zh) 一种处理方法、控制方法、识别方法及其装置和电子设备
JP2019117623A (ja) 音声対話方法、装置、デバイス及び記憶媒体
WO2017059815A1 (zh) 一种快速识别方法及家庭智能机器人
CN110459222A (zh) 语音控制方法、语音控制装置及终端设备
US11705129B2 (en) Voice control method, cloud server and terminal device
CN110767225B (zh) 一种语音交互方法、装置及系统
JP2021520558A (ja) ルーチンの実行中のクライアントデバイス同士の間の自動アシスタントルーチンの転送
WO2016078214A1 (zh) 终端处理方法、装置及计算机存储介质
CN112634897B (zh) 设备唤醒方法、装置和存储介质及电子装置
CN110111789A (zh) 语音交互方法、装置、计算设备和计算机可读介质
CN111161714A (zh) 一种语音信息处理方法、电子设备及存储介质
US20220028384A1 (en) Voice control method, cloud server and terminal device
CN111192590B (zh) 语音唤醒方法、装置、设备及存储介质
CN111261160B (zh) 一种信号处理方法及装置
CN108172228B (zh) 语音命令词更换方法、装置、声控设备及计算机存储介质
CN107742520B (zh) 语音控制方法、装置及系统
CN108806678A (zh) 设备的语音唤醒方法及装置
CN112908321A (zh) 设备控制方法、装置、存储介质及电子装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18920722

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18920722

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 170521)

122 Ep: pct application non-entry in european phase

Ref document number: 18920722

Country of ref document: EP

Kind code of ref document: A1