WO2018157388A1

WO2018157388A1 - Wake-up method and device for robot, and robot

Info

Publication number: WO2018157388A1
Application number: PCT/CN2017/075588
Authority: WO
Inventors: 骆磊
Original assignee: 深圳前海达闼云端智能科技有限公司
Priority date: 2017-03-03
Filing date: 2017-03-03
Publication date: 2018-09-07
Also published as: CN107223280A; CN107223280B

Abstract

A wake-up method for a robot, comprising: acquiring a voice command (101); parsing the wake-up words and command content in the voice command according to the voice command and a preconfigured wake-up word database (102), the wake-up word database comprising at least two wake-up words, and the wake-up words in the wake-up word database corresponding to at least two robots; and executing an operation according to the parsed wake-up words and command content (103). The method can correctly parse the command content in the voice command, thereby correctly completing a task assigned by a user.

Description

Robot wake-up method, device and robot

Technical field

Embodiments of the present invention relate to the field of artificial intelligence automatic control, for example, to a robot wake-up method, apparatus, and robot.

Background technique

With the development of artificial intelligence technology, robots bring a lot of convenience to human production and life. When the current user controls the robot, the robot can pre-set the wake-up words. When the robot hears a specific wake-up word, it knows that the user is calling himself. The user can issue a voice command to the robot in the form of a wake-up word plus command content, for example, "Mike (awake word), what is the weather today (command content)?", the robot that hears the voice command will parse the voice command. The wake-up word is parsed. If the wake-up word is consistent with the wake-up word set in the robot, the robot will be able to discriminate that it is calling itself, and the content after the wake-up word in the voice command is used as the command content, that is, the user is identified today. The weather, so you can directly answer, as shown in Figure 1 shows the user to control a robot.

In the process of implementing the present invention, the inventors have found that at least the following problems exist in the related art: when a same user owns a plurality of robots, and a certain task requires multiple robots to complete together, the existing method will no longer be applicable. For example, the user command: "Mike, Tom, Jerry, help me clean up the room", according to the above analysis method, the command content of the wake-up word for Mike's robot is "Tom, Jerry, help me clean up the room." Since the command content cannot be parsed correctly, the robot will not be able to complete the task that the user confesses.

Summary of the invention

An object of the embodiments of the present invention is to provide a new robot wake-up method, device, and robot. When a user simultaneously issues a voice command to multiple robots, the robot can correctly parse the command content, thereby correctly completing the task that the user confesses. .

In a first aspect, an embodiment of the present invention provides a robot wake-up method, where the wake-up method is applied to a robot, and the method includes:

Get a voice command;

And parsing the wake-up words and the command content in the voice command according to the voice command and the preset wake-up word library, the wake-up word library includes at least two wake-up words, and the wake-up words in the wake-up word library are used for Wake up at least two robots corresponding to one or more wake words;

Perform operations based on the parsed wakeup words and command content.

In a second aspect, an embodiment of the present invention further provides a robot wake-up device, where the wake-up device should For a robot, the device includes:

a voice command acquisition module, configured to acquire a voice command;

a voice command parsing module, configured to parse an awakening word and a command content in the voice command according to the voice command and a preset wake-up term library, where the wake-up word library includes at least two wake-up words, the wake-up word The wake-up words in the library are used to wake up at least two robots, the robot corresponding to one or more wake-up words;

An execution module is configured to perform an operation according to the parsed wake-up words and command content.

In a third aspect, an embodiment of the present invention further provides a robot, including:

At least one processor; and,

a memory communicatively coupled to the at least one processor; wherein

The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method as described above.

The wake-up method and device provided by the embodiment of the present invention can preset the wake-up words corresponding to the plurality of robots to the wake-up words by preset a wake-up word library including at least two wake-up words in the robot, for example, the user owns multiple robots. In the library, when the user issues a voice command to multiple robots at the same time, the robot can correctly parse the wake-up words contained in the voice command according to the preset wake-up vocabulary, thereby correctly parsing the command content in the voice command, and thus correct. Complete the task that the user confesses.

DRAWINGS

The one or more embodiments are exemplified by the accompanying drawings in the accompanying drawings, and FIG. The figures in the drawings do not constitute a scale limitation unless otherwise stated.

1 is a schematic diagram of a user controlling a robot in the prior art;

2 is a schematic diagram of an application scenario of the method and apparatus of the present invention;

3 is a flow chart of one embodiment of a wake-up method of the present invention;

4a is a flow chart showing steps of acquiring a voice command in an embodiment of the wake-up method of the present invention;

4b is a flowchart of a step of acquiring a voice command in an embodiment of the wake-up method of the present invention;

5 is a flow chart showing steps of parsing a voice command in an embodiment of the wake-up method of the present invention;

Figure 6a is a flow chart showing the steps of executing a voice command in one embodiment of the wake-up method of the present invention;

Figure 6b is a flow chart showing the steps of executing a voice command in one embodiment of the wake-up method of the present invention;

Figure 7a is a flow chart of one embodiment of a wake-up method of the present invention;

Figure 7b is a flow chart of one embodiment of the wake-up method of the present invention;

Figure 8 is a schematic structural view of an embodiment of the wake-up device of the present invention;

9 is a schematic structural diagram of a voice command acquiring module in an embodiment of the wake-up device of the present invention;

Figure 10 is a schematic structural view of an embodiment of the wake-up device of the present invention;

Figure 11 is a block diagram showing the structure of an embodiment of the wake-up device of the present invention;

FIG. 12 is a schematic diagram showing the hardware structure of a waking method according to an embodiment of the present invention.

detailed description

The technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. It is a partial embodiment of the invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

The robot wake-up method and apparatus provided by the present invention are applicable to an application scenario as shown in FIG. 2, and include a plurality of robots 20, which can communicate with each other through a network 30, wherein the network 30 can be, for example, a home or a company. LAN, or a specific network, etc. The bot 20 has at least one network interface that establishes a communication connection with the network 30 to retrieve data or instructions from the network 30. The user 10 can set or issue commands to the plurality of robots 20.

Each robot has its corresponding wake-up word for waking itself from the sleep state or responding to the user's call (the wake-up word is generally one or more). The wake-up word may be a name, an identifier or any other vocabulary of the robot, and the wake-up word may be set by the user or may be provided at the factory.

Each robot has a wake-up vocabulary for placing wake-up words, which can be shipped from the factory or set by the user. The same wake-up vocabulary can be shared between the robots of the same user. Taking the embodiment shown in Figure 2 as an example, assume that the user 10 has three robots, and the wake-up words of the three robots are Mike, Tom, and Jerry (here Each robot corresponds to a wake-up word as an example, but it is not limited to this. Each robot can also associate more than two wake-up words. Then, the wake-up vocabularies of the three robots can be set to include Mike, Tom, and Jerry. When the user simultaneously issues a voice command "Mike, Tom, Jerry, help me clean up the room" to the three robots, since each robot's wake-up vocabulary includes Mike, Tom, and Jerry, according to the wake-up vocabulary, the robot You can parse out the wake-up words Mike, Tom, and Jerry so that you can correctly parse the command content after the wake-up word "help me clean up the room."

In actual use, the acoustic model can be used to resolve the wake-up words in the voice command. The wake-up words in the wake-up lexicon use the phoneme sequence corresponding to the wake-up word, and the phoneme sequence is decoded according to the voice and the preset acoustic model. The phoneme sequence is then matched with the wake-up word phoneme sequence to resolve the wake-up word. After the wake-up word is parsed, the content behind the wake-up word in the voice command is used as the command content. The semantic analysis of the command content in the voice command requires the command lexical method file to be constructed in advance. The command content sent by the user needs to exist in the pre-built command lexical file, and the semantics of the command content is parsed according to the voice and command lexical file sent by the user.

The update of the wakeup vocabulary can be set by the user or can be done by the robot itself. For example, the current wake-up vocabulary includes three wake-up words: Mike, Tom, and Jerry. When the user purchases a robot (assuming the wake-up word is John) and establishes a communication connection with other robots through the network, the newly added robot will The other wake-up words John are broadcasted to other robots through the network. After receiving the wake-up words, other robots will add the wake-up words to the wake-up vocabulary and send the updated wake-up vocabulary to the newly added robot.

It should be noted that although only one user 10 and three robots 20 are shown in FIG. 2 . However, those skilled in the art can understand that the application scenario may further include more users 10 and robots 20 in the actual application process. The robot wake-up method and apparatus provided by the present invention are also applicable to a case where a user controls one or two robots.

An embodiment of the present invention provides a robot wake-up method, which may be performed by any of the robots shown in FIG. 2, as shown in FIG. 3, which is a flowchart of an embodiment of the wake-up method, and the wake-up method. include:

Step 101: Acquire a voice command.

In practical applications, a microphone can be placed on the robot for receiving voice signals in real time. The voice command may be a voice signal received in real time. However, sometimes the user makes a voice, but it is not necessarily a voice command for the robot. Therefore, it is necessary to further judge the voice information. If the voice message is a command for the robot corresponding to any wake-up word in the wake-up dictionary, Voice information as a voice command.

Step 102: Parsing an awakening word and a command content in the voice command according to the voice command and a preset wake-up term library, where the wake-up word library includes a plurality of wake-up words, and the wake-up words in the wake-up word library Corresponding to at least two robots;

The command content may be a voice corresponding to the command content intercepted from the voice command sent by the user, or may be a result of semantically parsing the voice. If it is the former, the robot needs to perform semantic analysis when performing the operation corresponding to the command content. The plurality of ones include one or more, that is, the wake-up vocabulary may include one or more wake-up words.

Step 103: Perform an operation according to the parsed wake-up word and the command content.

Optionally, after the robot parses the wake-up word and the command content, if the wake-up word in the parsing includes the corresponding wake-up word, the operation corresponding to the command content is performed, and if the corresponding wake-up word is not included, no operation is performed. Alternatively, regardless of whether the parsed wake-up word contains its own corresponding wake-up word, the command content is sent to other robots corresponding to the wake-up word, and the robot corresponding to the wake-up word completes the task that the user confesses. The robot may also decompose the task that the user confesses into several subtasks according to the command content, notify the subtasks of the other robots corresponding to the wakeup words, and coordinate the tasks corresponding to the user by the robot corresponding to the wakeup words.

It should be noted that the

above steps

101, 102, and 103 are not necessarily performed by each robot, and it is possible for any robot to perform all steps or only one or two of them. Figure 2 The application scenario shown is an example. When the user makes a voice "Mike, Tom, Jerry, help me clean up the room". The robots Mike, Tom, and Jerry can respectively obtain the voice commands issued by the user, and after parsing the wake-up words and the command content, if they find that they are the command objects of the user, the operations are performed according to the parsed command content. In this case, all three robots perform

steps

101, 102, and 103. It is also possible that only the robot Mike hears the voice command issued by the user. After Mike obtains the voice command, Mike parses out the wake-up word and the command content, and sends the command content to Tom and Jerry, and the three robots perform the task together. In this case only Mike performed

steps

101, 102 and 103, while Tom and Jerry only performed step 103. Step 102 may be performed by only one robot or several robots, and the executed robot will share the parsed wake-up words and command content to other robots through the network.

The wake-up method provided by the embodiment of the present invention can preset the wake-up words corresponding to the plurality of robots in the wake-up word library by preset a wake-up word library including a plurality of wake-up words in the robot, for example, the user owns multiple robots. When the user issues a voice command to multiple robots at the same time, the robot can correctly parse the wake-up words contained in the voice command according to the preset wake-up vocabulary, thereby correctly parsing the command content in the voice command, thereby completing the user's account. task.

Specifically, in some embodiments of the present invention, as shown in FIG. 4a, the acquiring a voice command includes the following steps;

Step 1011: Listening to voice information;

Each robot will listen to the voice information sent by the user in real time;

Step 1012: Confirm whether the voice information is a command for a robot corresponding to the wakeup vocabulary;

Step 1013: If the voice information is a command for the robot corresponding to the wakeup vocabulary, the voice information is recorded as a voice command.

It is confirmed whether the voice information is a command issued by the user for the robot corresponding to any wake-up word in the wake-up dictionary, and if yes, the voice information is a voice command issued by the user. Specifically, whether the voice information is a command for the robot corresponding to the wakeup vocabulary, first confirm whether the voice information includes any wake-up words in the preset wake-up vocabulary, if any wake-up in the wake-up vocabulary is included The word continues to determine whether the appearance of the wake-up word is a user's command or call to the robot, rather than the user mentioning the robot in a call with other people. Determining whether the voice information sent by the user is a command or a call to the robot, and determining whether the time interval between the wake-up word and the following voice content is more than a preset time. If the preset time is exceeded, the voice information is Command or call to the robot. Or by judging whether there is other voice content in front of the first wake-up word, if there is no other voice content, the voice information is an order or call to the robot.

Optionally, in another embodiment of the waking method, as shown in FIG. 4b, the acquiring a voice command includes:

Step 1014: Listening to voice information;

Each robot will listen to the voice information sent by the user in real time;

Step 1015: Confirm whether the voice information is a command for a robot corresponding to the wakeup vocabulary;

Step 1016: If the voice information is a command for the robot corresponding to the wakeup vocabulary, record the voice information and the start time of the voice information, and add a preset temporary command record group;

If the robot monitors the voice information sent by the user, it further determines whether the voice information is a command issued by the user for the robot corresponding to any wake-up word in the wake-up dictionary, and if so, records the voice information and the robot hears The moment of the voice message, and join the preset temporary command record group.

In practical applications, the robot can store the voice information and the moment when the voice information is heard into the cache of the robot. The purpose of establishing a temporary command group is to obtain relatively complete and clear user commands. Take the three robots shown in Figure 2 as an example. Suppose the voice command issued by the user is “Mike, Tom, Jerry, help me clean up the room”. Then every robot that hears the beginning of "Mike", the beginning of "Tom" or the beginning of "Jerry" records the timestamp t (starting time) when it hears the above-mentioned wake-up words, and not necessarily every robot can hear the complete The order, maybe the robot is moving to the user in the distance, did not hear the initial "Mike", heard "Tom, Jerry, help me clean up the room", the timestamp recorded here is to prevent This half-statement occurs as a complete command.

Specifically, whether the voice information is a command for the robot corresponding to the wakeup vocabulary, first confirm whether the voice information includes any wake-up words in the preset wake-up vocabulary, if any wake-up in the wake-up vocabulary is included The word continues to determine whether the appearance of the wake-up word is a user's command or call to the robot, rather than the user mentioning the robot in a call with other people. Determining whether the voice information sent by the user is a command or a call to the robot, and determining whether the time interval between the wake-up word and the following voice content is more than a preset time. If the preset time is exceeded, the voice information is Command or call to the robot. Or by judging whether there is other voice content in front of the first wake-up word, if there is no other voice content, the voice information is an order or call to the robot.

Step 1017: Confirm the earliest time in the start time of the robot record in the temporary command record group, and determine the start time period according to the earliest time;

The starting time of the robot record in the temporary command record group is compared, and the earliest time is determined because the voice information recorded at the earliest time is relatively complete. For example, the earliest time is t1, the earliest time t1 is taken as the starting point, and the starting time period is determined from t1 to t1+t0 according to an empirical setting threshold t0.

The setting of t0 can be set according to the performance and experience of the robot, for example, 0.1s. On the one hand, the response of different robots is fast and slow, so it is necessary to set a time difference. On the other hand, there is a relatively fixed interval between the two wake-up words sent by the user, so the set time difference should be less than this interval. In case of missing a wake-up word.

Step 1018: Obtain the voice information with the highest resolution among the voice information whose starting time is located in the starting time period as a voice command.

After the voice information recording is finished, the robot in the temporary command recording group will judge the clarity of the voice information buffered by itself, and obtain the score value x, and take the highest-resolution voice information in the starting time period (t1 to t1+t0). As a voice command.

It should be noted that the

above steps

1014, 1015, 1016, 1017, and 1018 are not necessarily performed by each robot. Generally, the robot that hears the voice information performs

steps

1014, 1015, and 1016, but steps 1017 and 1018 may only Executed by one robot or several robots, each robot can broadcast its own working state to other robots, which is executed by the most idle robot, and then the executed robot will share the execution results to other robots through the network.

In the embodiment of the present invention, each robot records the start time of the voice information, and joins the temporary command record group. By comparing the start time of the robot record in the temporary command record group, the earliest time when the voice information is heard can be determined. Get a relatively complete voice command. By determining the speech information with the highest resolution in the initial time period as the voice command, relatively complete and clear voice commands can be obtained, and the reliability of acquiring the user voice command is enhanced. The clarity of the voice command facilitates the correct interpretation of the voice command, and the integrity of the voice command can ensure that the wake-up words sent by the user can be parsed as completely as possible, and the cooperation between the robots corresponding to the wake-up words can be facilitated.

Specifically, in some embodiments of the waking method, as shown in FIG. 5, the wakeup word and the command content in the voice command are parsed according to the voice command and the preset wakeup vocabulary. include:

Step 1021: Parse the wake-up words in the voice command according to the voice command and the wake-up vocabulary;

Since the user may command one or several commands at the same time, the grammatical format of the user's voice command to the robot is generally as follows:

<name>, [and | and | and | that | um | ...] <name>, [and | and | and | that | ah | ...] <name>, ..., <command>.

Where <name> is the wake-up word of a certain robot, and the number of occurrences of <name> may be one or more;

[And | and | and | that | ah | ...] are two possible conjunctions between the awakening words, such as "Zhang San and Li Si", this conjunction does not necessarily appear, so use [] to indicate;

<command> is the content of the command after the wake-up word.

For example, the user voice command is "Zhang San, Li Si, and Wang Wu, help me prepare lunch before twelve o'clock." This sentence can be applied according to the above grammatical format, then it can be judged "Zhang San, Li Si, Wang Wu "The awakening words for the three robots, "also" is a conjunction or a spoken word, "help me prepare lunch before twelve o'clock" as the command content.

By matching the voice command and the wake-up words in the wake-up vocabulary one by one, the wake-up words in the voice command can be parsed.

Step 1022: Parse the command content in the voice command according to the voice command and the wake-up word.

After the wake-up word in the voice command is parsed, the content behind the wake-up word in the voice command is the actual command content of the user, and the semantic meaning of the command content can be used to know the user's true intention.

It should be noted that the

above steps

1021 and 1022 are not necessarily performed by each robot, and may be performed by only one robot or several robots. Each robot can broadcast its own working state to other robots, and the most idle robots. To execute.

Specifically, in some embodiments of the waking method, as shown in FIG. 6a, the performing operations according to the parsed wake-up words and command content includes:

Step 1031: Notifying the robot corresponding to the wake-up word with the command content, so that the robot corresponding to the wake-up word performs an operation corresponding to the command content.

The robot that can parse the wake-up word and the command content may also be a robot that acquires the wake-up word and the command content through the network to notify the command content to the wake-up word. For example, the voice command issued by the user is "Mike, Tom, Jerry, help me clean up the room." If the robot that parses the voice command is Mike, then Mike will send the command content "Help me clean up the room" to Tom and Jerry, then Mike, Tom and Jeery clean up the room together. Among them, the command content that Mike sends to Tom and Jeery can be the result of semantic analysis, or it can be voice, and Tom and Jerry need to perform semantic analysis on the voice.

Optionally, in another embodiment of the waking method, as shown in FIG. 6b, the performing operations according to the parsed wake-up words and command content, including:

Step 1032: According to the command content decomposition task, respectively notify the decomposed task to the robot corresponding to the wake-up word, so that the robot corresponding to the wake-up word cooperates to perform the operation corresponding to the command content.

In practical applications, after the robot parses the wake-up word and the command content, the robot that parses the wake-up word and the command content may also be another robot that acquires the wake-up word and the command content through the network, and may notify other robots corresponding to the wake-up word to wait. Tasks, so that each robot establishes a task group, and the task group members start to synchronize (such as sharing location information, unique identifier, own ability, etc.), and then wait for the task. The robot that parses the voice command combines the position of the robot and its own capabilities according to the command content, disassembles the task that the user confesses, divides it into several subtasks, and then sends the disassembled subtask to other members in the task group, the task group. The robots within the collaboration cooperate to complete the tasks that the user confesses. For example, taking Figure 2 as an example, assuming that the robot that parses the voice command is Mike, Mike divides the task of cleaning up the room into three subtasks: cleaning up the living room and bedroom, cleaning up the bathroom, and cleaning up the kitchen. Mike can clean up the bathroom and clean up the kitchen. The missions were sent to Tom and Jerry to clean up the living room and bedroom.

Compared with the prior art, the main body of the task is no longer a separate robot, but a plurality of robots are combined to perform tasks, and the execution efficiency is higher and the user experience is good.

As shown in FIG. 7a, it is a schematic flowchart of an embodiment of the waking method. In this embodiment, the waking method includes:

Step 201: Listening to voice information;

Each robot will listen to the voice information sent by the user in real time;

Step 202: Confirm whether the voice information is a command for a robot corresponding to the wakeup vocabulary;

Step 203: If the voice information is a command for the robot corresponding to the wakeup vocabulary, the voice information is recorded as a voice command.

Each robot that listens to the voice information confirms whether the voice information is a command issued by the user for the robot corresponding to any wake-up word in the wake-up vocabulary, and if so, the voice information is used as a voice command.

Step 204: Parse the wake-up word and the command content in the voice command according to the voice command and the preset wake-up dictionary.

In this embodiment, each robot that listens to a voice command parses the voice command and parses the wake word and the command content.

Step 205: Notifying the robot corresponding to the wake-up word with the command content, so that the robot corresponding to the wake-up word performs an operation corresponding to the command content.

The robot that parses the voice command, if the parsed wake-up word does not include itself, sends the command content to the robot corresponding to the other wake-up words; if the parsed wake-up word includes itself, starts executing the operation corresponding to the command content, and the command The content is sent to other related robots. If the robot parses the command content itself, the operation is performed according to the command content that is parsed by itself, and if the command content is not parsed by itself, the operation is performed according to the command content sent by the other robot.

According to the wake-up method provided by the embodiment of the present invention, the robot can correctly parse the wake-up words contained in the voice command according to the preset wake-up vocabulary by preset a wake-up vocabulary including a plurality of wake-up words in the robot, thereby correctly analyzing The content of the command in the voice command is completed, and the task that the user confesses is correctly completed. The user can issue commands to multiple robots at the same time, and multiple robots can jointly perform the tasks issued by the user, and the execution efficiency is higher and the user experience is good.

As shown in FIG. 7b, it is a schematic flowchart of an embodiment of the waking method. In this embodiment, the waking method includes:

Step 301: Listening to voice information;

Each robot will listen to the voice information sent by the user in real time;

Step 302: Confirm whether the voice information is a command for a robot corresponding to the wakeup vocabulary;

Step 303: If the voice information is a command for the robot corresponding to the wakeup vocabulary, record the voice information and the start time of the voice information, and add a preset temporary command record group;

Each robot that listens to the voice information confirms whether the voice information is a command issued by the user for the robot corresponding to any wake-up word in the wake-up dictionary, and if so, records the voice information and listens to the start of the voice information. At the moment, and join the preset temporary command record group. Then, the robot that joins the temporary command record group will broadcast the temporary command record group to other robots to Make the members of the temporary command record group aware of the existence of other group members.

Step 304: Confirm the earliest time in the start time of the robot record in the temporary command record group, and determine the start time period according to the earliest time;

The robot in the temporary command record group broadcasts the start time of its own record to other robots in the group, and a certain robot determines the earliest time in each start time and determines the start time period based on the earliest time.

Step 305: Obtain voice information with the highest resolution among the voice information whose starting time is located in the starting time period as a voice command.

After the voice information recording is finished, the robot in the temporary command recording group performs the resolution determination on the voice information recorded by itself, obtains the sharpness score value, and then broadcasts the sharpness score value to other robots in the temporary command record group, a certain robot It will find the highest-resolution voice information in the starting time period.

Step 306: Parsing the wake-up words and the command content in the voice command according to the voice command and the preset wake-up term library;

Step 307: According to the command content decomposition task, respectively notify the decomposed task to the robot corresponding to the wake-up word, so that the robot corresponding to the wake-up word cooperates to perform the operation corresponding to the command content.

It should be noted that the

above steps

304, 305, 306, and 307 are not necessarily performed by each robot, and each robot can broadcast its own working state to other robots, and is executed by the most idle robot.

According to the wake-up method provided by the embodiment of the present invention, the robot can correctly parse the wake-up words contained in the voice command according to the preset wake-up vocabulary by preset a wake-up vocabulary including a plurality of wake-up words in the robot, thereby correctly analyzing The content of the command in the voice command is completed, and the task that the user confesses is correctly completed. By recording the start time of the voice information monitored by each robot and adding a temporary command record group, by comparing the start time of the robot record in the temporary command record group, the earliest time to hear the voice information can be determined, and a relatively complete can be obtained. Voice command. By determining the speech information with the highest resolution in the initial time period as the voice command, relatively complete and clear voice commands can be obtained, and the reliability of acquiring the user voice command is enhanced. The user can issue commands to multiple robots at the same time, and multiple robots can complete the tasks issued by the users in a coordinated manner, and the execution efficiency is higher and the user experience is good.

Optionally, in another embodiment of the waking method, the method further includes:

Update the wakeup vocabulary.

When the user purchases another robot and establishes a communication connection with other robots through the network, the newly added robot will broadcast its wake-up words to other robots through the network, and the robot receiving the wake-up words will synchronize the wake-up words to The own wake-up vocabulary, and send the synchronized corpus of wake-up vocabulary back to the newly added robot.

Dismiss the temporary command record group.

After the task assignment is completed, each robot starts to perform the task. At this time, the temporary command record group can be dismissed, the memory is released, and the resource utilization rate is improved.

Correspondingly, the embodiment of the present invention further provides a robot wake-up device, which is disposed in any of the robots shown in FIG. 2. As shown in FIG. 8, the wake-up device 400 includes:

The voice command obtaining module 401 is configured to acquire a voice command.

The voice command parsing module 402 is configured to parse the wake-up word and the command content in the voice command according to the voice command and the preset wake-up vocabulary, where the wake-up vocabulary includes a plurality of wake-up words, the wake-up word The wake-up words in the library correspond to at least two robots;

The executing module 403 is configured to perform an operation according to the parsed wake-up word and the command content.

The wake-up device provided by the embodiment of the present invention can correct the wake-up words contained in the voice command according to the preset wake-up vocabulary by presetting a wake-up vocabulary including a plurality of wake-up words in the robot, thereby correctly analyzing The content of the command in the voice command is completed, and the task that the user confesses is correctly completed.

Specifically, in some embodiments of the waking device, the voice command obtaining module 401 includes:

a voice information monitoring sub-module for monitoring voice information;

a voice command confirmation submodule, configured to confirm whether the voice information is a command for a robot corresponding to the wakeup vocabulary;

The first voice command acquisition submodule is configured to record the voice information as a voice command if the voice information is a command for a robot corresponding to the wakeup vocabulary.

Optionally, in other embodiments of the waking device, as shown in FIG. 9, the voice command obtaining module 401 includes:

a voice information monitoring sub-module 4011, configured to monitor voice information;

a voice command confirmation sub-module 4012, configured to confirm whether the voice information is a command for a robot corresponding to the wake-up vocabulary;

The voice command recording sub-module 4013 is configured to record the voice information and the start time of the voice information if the voice information is a command for the robot corresponding to the wake-up vocabulary, and add a preset temporary command record group. ;

a start time period confirmation sub-module 4014, configured to confirm an earliest time in a start time of the robot record in the temporary command record group, and determine a start time period according to the earliest time;

The second voice command acquisition sub-module 4015 is configured to obtain voice information with the highest resolution among the voice information whose starting time is located in the start time period as a voice command.

Specifically, in some embodiments of the wake-up device, the voice command confirmation sub-module includes:

a voice command confirmation subunit, configured to: if the voice information includes any one of a preset wakeup vocabulary The wake-up word and the appearance of the wake-up word are called, and the voice information is a command for the robot corresponding to the wake-up vocabulary.

Optionally, as shown in FIG. 10, in other embodiments of the awake device, the awake device 500 includes: a voice command obtaining module 501, a voice command parsing module 502, and an executing module 503,

The wakeup vocabulary update module 504 is configured to update the wakeup vocabulary.

In some embodiments, the wakeup vocabulary update module includes:

The first wakeup vocabulary update submodule is configured to set a wakeup word according to the wakeup word setting instruction, and broadcast the wakeup word.

In other embodiments, the wakeup vocabulary update module further includes:

The second wake-up vocabulary update sub-module is configured to receive the broadcasted wake-up word, add the wake-up word to the preset wake-up vocabulary, and send the updated wake-up vocabulary to the robot that broadcasts the wake-up word.

Specifically, in some embodiments of the waking device, the voice command parsing module includes:

The wake-up word parsing sub-module is configured to parse the wake-up words in the voice command according to the voice command and the wake-up vocabulary;

The command content parsing sub-module is configured to parse the command content in the voice command according to the voice command and the wake-up word.

Specifically, in some embodiments of the waking device, the execution module includes:

The first execution submodule is configured to notify the robot corresponding to the wakeup word of the command content, so that the robot corresponding to the wakeup word performs an operation corresponding to the command content.

In other embodiments of the wake-up device, the execution module includes:

The second execution sub-module is configured to respectively notify the robot corresponding to the wake-up word according to the command content decomposition task, so that the robot corresponding to the wake-up word cooperates to perform the operation corresponding to the command content.

FIG. 11 is a schematic structural diagram of an embodiment of the wake-up device. In this embodiment, the wake-up device 600 includes:

The voice command obtaining module 601 is configured to obtain a voice command. The voice command acquiring module 601 includes:

a voice information monitoring sub-module 6011, configured to monitor voice information;

a voice command confirmation sub-module 6012, configured to confirm whether the voice information is a command for a robot corresponding to the wake-up vocabulary;

The voice command recording sub-module 6013 is configured to record the voice information and the start time of the voice information if the voice information is a command for the robot corresponding to the wake-up vocabulary, and add a preset temporary command record group. ;

a start time period confirmation sub-module 6014 for confirming the robot record in the temporary command record group The earliest time in the starting time, determining the starting time period according to the earliest time;

The second voice command acquisition sub-module 6015 is configured to obtain voice information with the highest resolution among the voice information whose starting time is located in the start time period as a voice command.

The voice command parsing module 602 is configured to parse the wake-up word and the command content in the voice command according to the voice command and the preset wake-up term library, where the wake-up word is used to wake up the robot; wherein the voice command is The parsing module 602 includes:

The wake-up word parsing sub-module 6021 is configured to parse the wake-up words in the voice command according to the voice command and the wake-up vocabulary;

The command content parsing sub-module 6022 is configured to parse the command content in the voice command according to the voice command and the wake-up word.

The executing module 603 is configured to perform an operation according to the parsed wake-up word and the command content, where the executing module 603 includes:

The second execution sub-module 6031 is configured to separately notify the robot corresponding to the wake-up word according to the command content decomposition task, so that the robot corresponding to the wake-up word cooperates to perform the operation corresponding to the command content.

The wakeup vocabulary update module 604 is configured to update the wakeup vocabulary.

The voice information monitoring sub-module 6011 monitors the voice sent by the user in real time, and the voice command confirmation sub-module 6012 confirms whether the voice information monitored by the voice information monitoring sub-module 6011 is a command of the user for the robot corresponding to the wake-up vocabulary, if the voice information is For the command of the robot corresponding to the vocabulary, the voice command recording sub-module 6013 records the voice information and the start time of the voice information, and adds a preset temporary command record group. The robot in the temporary command record group broadcasts the start time of the record to other robots, and the start time period confirmation sub-module 6014 confirms the earliest time to listen to the user voice command according to each start time, and determines the start according to the earliest time. period. After the voice information recording is finished, the robot in the temporary command recording group performs the resolution determination on the voice information recorded by itself, obtains the sharpness score value, and then broadcasts the sharpness score value to the other robots in the temporary command record group, the second voice. The command acquisition sub-module 6015 determines the voice information with the highest resolution among the voice information whose start time is located in the start time period as a voice command. The wake-up word analysis sub-module 6021 parses the wake-up word according to the voice command, and the command content analysis sub-module 6022 parses the command content according to the voice command and the wake-up word. The second execution sub-module 6031 causes the robot corresponding to the wake-up word to cooperate to perform an operation corresponding to the command content according to the command content decomposition task.

The wake-up device provided by the embodiment of the present invention can correct the wake-up words contained in the voice command according to the preset wake-up vocabulary by presetting a wake-up vocabulary including a plurality of wake-up words in the robot, thereby correctly analyzing The content of the command in the voice command is completed, and the task that the user confesses is correctly completed. The start time of the voice information is monitored by each robot, and the temporary command record group is added, and the voice information can be determined by comparing the start time of the robot record in the temporary command record group. At the earliest moment, you can get relatively complete voice commands. By determining the speech information with the highest resolution in the initial time period as the voice command, relatively complete and clear voice commands can be obtained, and the reliability of acquiring the user voice command is enhanced. The user can issue commands to multiple robots at the same time, and multiple robots can complete the tasks issued by the users in a coordinated manner, and the execution efficiency is higher and the user experience is good.

It should be noted that the above-mentioned wake-up device can perform the wake-up method provided by the embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. For a technical detail that is not described in detail in the wake-up device embodiment, reference may be made to the wake-up method provided by the embodiment of the present invention.

FIG. 12 is a schematic diagram showing the hardware structure of the robot 700 for the robot wake-up method according to the embodiment of the present invention. As shown in FIG. 12, the robot 700 includes:

One or more processors 701 and memory 702, one processor 701 is taken as an example in FIG.

The processor 701 and the memory 702 may be connected by a bus or other means, as exemplified by a bus connection in FIG.

The memory 702 is a non-volatile computer readable storage medium, and can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as program instructions corresponding to the wake-up method in the embodiment of the present invention. The module (for example, the voice command acquisition module 401, the voice command analysis module 402, and the execution module 403 shown in FIG. 8). The processor 701 executes various functional applications of the server and data processing by executing non-volatile software programs, instructions, and modules stored in the memory 702, that is, implementing the wake-up method of the above method embodiments.

The memory 702 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function; the storage data area may store data created according to the use of the wake-up device, and the like. Moreover, memory 702 can include high speed random access memory, and can also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, memory 702 can optionally include memory remotely located relative to processor 701 that can be connected to the wake-up device over a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The one or more modules are stored in the memory 702, and when executed by the one or more processors 701, perform the wake-up method in any of the above method embodiments, for example, performing the above described FIG. Method step 101 to step 103, method step 1011 to step 1013 in Fig. 4a, method step 1014 to step 1018 in Fig. 4b, method step 1021 to step 1022 in Fig. 5, method step 1031 in Fig. 6a, Fig. 6b Method step 1032, method steps 201-205 in FIG. 7a, method steps 301-307 in FIG. 7b; implementing modules 401-403 in FIG. 8, sub-modules 4011 and 4015 in FIG. 9, module 501 in FIG. - 504, the functions of the modules 601-604, the sub-module 6011-6015, the sub-module 6021-6022, and the sub-module 6031 in FIG.

The above product can perform the method provided by the embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. For technical details that are not described in detail in this embodiment, reference may be made to the embodiments of the present invention. The method provided.

Embodiments of the present invention provide a non-transitory computer readable storage medium storing computer-executable instructions that are executed by one or more processors, such as in FIG. The processor 701 is configured to enable the one or more processors to perform the wake-up method in any of the foregoing method embodiments, for example, to perform the method steps 101 to 103 in FIG. 3 described above, the method steps in FIG. 4a 1011 to step 1013, method step 1014 to step 1018 in Fig. 4b, method step 1021 to step 1022 in Fig. 5, method step 1031 in Fig. 6a, method step 1032 in Fig. 6b, method step 201 in Fig. 7a - 205, method steps 301-307 in FIG. 7b; implementing modules 401-403 in FIG. 8, sub-modules 4011 and 4015 in FIG. 9, modules 501-504 in FIG. 10, modules 601-604, sub-modules in FIG. Functions of 6011-6015, submodule 6021-6022, and submodule 6031.

The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, ie may be located A place, or it can be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment.

Through the description of the above embodiments, those skilled in the art can clearly understand that the various embodiments can be implemented by means of software plus a general hardware platform, and of course, by hardware. A person skilled in the art can understand that all or part of the process of implementing the above embodiments can be completed by a computer program to instruct related hardware, and the program can be stored in a computer readable storage medium. When executed, the flow of an embodiment of the methods as described above may be included. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not limited thereto; in the idea of the present invention, the technical features in the above embodiments or different embodiments may also be combined. The steps may be carried out in any order, and there are many other variations of the various aspects of the invention as described above, which are not provided in the details for the sake of brevity; although the invention has been described in detail with reference to the foregoing embodiments, It should be understood by those skilled in the art that the technical solutions described in the foregoing embodiments may be modified or equivalently substituted for some of the technical features; and the modifications or substitutions do not deviate from the embodiments of the present invention. The scope of the technical solution.

Claims

A robot wake-up method, the wake-up method applied to a robot, wherein the method comprises:

Get a voice command;

And parsing the wake-up words and the command content in the voice command according to the voice command and the preset wake-up word library, the wake-up word library includes at least two wake-up words, and the wake-up words in the wake-up word library are used for Wake up at least two robots corresponding to one or more wake words;

Perform operations based on the parsed wakeup words and command content.
The method according to claim 1, wherein the obtaining the voice command comprises:

Monitor voice information;

Determining whether the voice information is a command for a robot corresponding to the wakeup vocabulary;

If the voice information is a command for a robot corresponding to the wakeup vocabulary, the voice information is recorded as a voice command.
The method according to claim 1, wherein the obtaining the voice command comprises:

Monitor voice information;

Determining whether the voice information is a command for a robot corresponding to the wakeup vocabulary;

If the voice information is a command for the robot corresponding to the wakeup vocabulary, recording the voice information and monitoring the start time of the voice information, and adding a preset temporary command record group;

Confirming the earliest time in the start time of the robot record in the temporary command record group, and determining the start time period according to the earliest time;

The speech information with the highest resolution among the speech information whose starting time is located in the initial time period is obtained as a voice command.
The method according to claim 2 or 3, wherein the confirming whether the voice information is a command for a robot corresponding to the wakeup vocabulary comprises:

If the voice information includes any wake-up word in the preset wake-up vocabulary and the appearance of the wake-up word is a call, the voice information is a command for the robot corresponding to the wake-up vocabulary.
The method according to any one of claims 1 to 4, further comprising:

Update the wakeup vocabulary;

The update wake-up vocabulary includes:

The wake-up word is set according to the wake-up word setting instruction, and the wake-up word is broadcast.
The method of claim 5, wherein the updating the wake-up vocabulary further comprises:

Receiving a wake-up word of the broadcast, adding the wake-up word to the preset wake-up word library, and updating the Wake up the thesaurus to the robot that broadcasts the wake-up words.
The method according to any one of claims 1 to 4, wherein the performing the operation according to the parsed wake-up word and the command content comprises:

The robot corresponding to the wake-up word is notified of the command content so that the robot corresponding to the wake-up word performs an operation corresponding to the command content.
The method according to any one of claims 1 to 4, wherein the performing the operation according to the parsed wake-up word and the command content comprises:

According to the command content decomposition task, the decomposed task is separately notified to the robot corresponding to the wake-up word, so that the robot corresponding to the wake-up word cooperates to perform the operation corresponding to the command content.
A robot wake-up device, the wake-up device being applied to a robot, characterized in that the device comprises:

a voice command acquisition module, configured to acquire a voice command;

a voice command parsing module, configured to parse an awakening word and a command content in the voice command according to the voice command and a preset wake-up term library, where the wake-up word library includes at least two wake-up words, the wake-up word The wake-up words in the library are used to wake up at least two robots, the robot corresponding to one or more wake-up words;

An execution module is configured to perform an operation according to the parsed wake-up words and command content.
The device according to claim 9, wherein the voice command acquisition module comprises:

a voice information monitoring sub-module for monitoring voice information;

a voice command confirmation submodule, configured to confirm whether the voice information is a command for a robot corresponding to the wakeup vocabulary;

The first voice command acquisition submodule is configured to record the voice information as a voice command if the voice information is a command for a robot corresponding to the wakeup vocabulary.
The device according to claim 9, wherein the voice command acquisition module comprises:

a voice information monitoring sub-module for monitoring voice information;

a voice command confirmation submodule, configured to confirm whether the voice information is a command for a robot corresponding to the wakeup vocabulary;

a voice command recording submodule, configured to record the voice information and the start time of the voice information if the voice information is a command for the robot corresponding to the wakeup vocabulary, and add a preset temporary command record group;

a start time period confirmation submodule, configured to confirm an earliest time in a start time of the robot record in the temporary command record group, and determine a start time period according to the earliest time;

a second voice command acquisition submodule, configured to obtain a language whose starting time is within the starting time period The highest-resolution voice information in the voice information is used as a voice command.
The device according to claim 10 or 11, wherein the voice command confirmation submodule comprises:

The voice command confirmation sub-unit is configured to: if the voice information includes any wake-up words in the preset wake-up vocabulary and the appearance of the wake-up word is a call, the voice information is a command for the robot corresponding to the wake-up vocabulary.
The device according to any one of claims 9 to 12, wherein the device further comprises:

The wakeup vocabulary update module is used to update the wakeup vocabulary;

The wakeup vocabulary update module includes:

The first wakeup vocabulary update submodule is configured to set a wakeup word according to the wakeup word setting instruction, and broadcast the wakeup word.
The device according to claim 13, wherein the wake-up thesaurus update module further comprises:

The second wake-up vocabulary update sub-module is configured to receive the broadcasted wake-up word, add the wake-up word to the preset wake-up vocabulary, and send the updated wake-up vocabulary to the robot that broadcasts the wake-up word.
The apparatus according to any one of claims 9 to 12, wherein the execution module comprises:

The first execution submodule is configured to notify the robot corresponding to the wakeup word of the command content, so that the robot corresponding to the wakeup word performs an operation corresponding to the command content.
The apparatus according to any one of claims 9 to 12, wherein the execution module comprises:

The second execution sub-module is configured to respectively notify the robot corresponding to the wake-up word according to the command content decomposition task, so that the robot corresponding to the wake-up word cooperates to perform the operation corresponding to the command content.
A robot characterized by comprising:

At least one processor; and,

a memory communicatively coupled to the at least one processor; wherein

The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method of any of claims 1-8 method.
A non-transitory computer readable storage medium, wherein the computer readable storage medium stores computer executable instructions that, when executed by a robot, cause the robot to execute an execution claim The method of any of 1-8.
A computer program product, characterized in that the computer program product comprises A computer program on a non-transitory computer readable storage medium, the computer program comprising program instructions that, when executed by a robot, cause the robot to perform the method of any of claims 1-8.