SYSTEM AND METHOD FOR DETERMINING THE TIMING OF A RESPONSE IN GROUP COMMUNICATION USING ARTIFICIAL INTELLIGENCE

Abstract

Methods (400) for receiving information, comprising: forming a conversation group (220) from a large number of radios (210); Sending a query via a (push-to-talk) PTT radio of the talkgroup (220) to at least one member of the talkgroup (220); and Generating and sending a response to a query from a KL server (240) with natural language processing and response capability after a predetermined response waiting time for the conversation group (220) has elapsed without a response being generated by the conversation group (220).

Description

Related registrations

[0001] This application is related to the US application with publication number US 2018 / 0181656 A1 and the US patent US 9,961,516 B1, which were jointly transferred to and are owned by Motorola Solutions, Inc. Field of invention

[0002] The present invention relates generally to radio communication systems and in particular to the application of artificial intelligence (Cl) in radio communication systems. background

[0003] Obtaining the best possible search results for every query via a virtual assistant, without disrupting communication, is crucial not only for the user of the search device but also for managing the entire communication system. A server used for virtual assistant queries and responses can handle numerous queries sent simultaneously by various devices. Efficient use of channel bandwidth and response timing, along with the ability to obtain the most accurate and up-to-date answer, are critical factors to consider when managing virtual assistant search queries within a communication system that offers such a search capability.Systems that have had limited ability to utilize the search capabilities of virtual assistants, such as public safety communication systems, could consider integrating such search capabilities if improvements are made to the system's handling capabilities. Such improvements would benefit not only public safety applications but any communication system that includes the ability to query virtual assistants. A method and associated system are already known from the prior art, exemplified by US 2015 / 0148084 A1, in which a voice message is analyzed by user equipment on a network using speech recognition, a recipient is identified, and the message is then forwarded to the recipient on the network.US 2008 / 0045256 A1 further describes a push-to-talk function on a mobile phone in which a conversation group is established by speaking the recipient's name. A speech recognition device located in the mobile phone or a push-to-talk server can recognize the recipient's name, determine the correct address for the message, establish a push-to-talk session, and transmit the message to the intended recipient. If the recipient is not online, an automatic reply is generated. Furthermore, prior art EP 1 643 708 A1 discloses a method and an associated system for using a push-to-talk connection in communication conferences managed by at least one non-human participant. In this process, voice input from participants is analyzed, and floor time is permitted if necessary.

[0004] Accordingly, there is a need to improve the handling of virtual search options within a communication system. Brief description of the characters

[0005] The accompanying illustrations, in which the same reference numerals refer to identical or functionally similar elements in the individual views, are included in the description together with the following detailed description and form part of it, serving to further illustrate embodiments and concepts that include the claimed invention and explain various principles and advantages of these embodiments. Fig. Figure 1A is a block diagram of the communication system, which is designed and functions according to some embodiments. Fig. 1B is a flowchart for handling the fullness of a response, according to some embodiments. Fig.Figure 2 is a block diagram of the communication system, which is designed and functions according to some embodiments. Fig. Figure 3 is a communication exchange diagram, according to some embodiments. Fig. Figure 4 is a flowchart for obtaining supplementary information, according to some embodiments. Fig. Figure 5 is a flowchart for a timing procedure, according to some embodiments.

[0006] Experts will recognize that elements in the figures are illustrated for the sake of simplicity and clarity and are not necessarily drawn to scale. For example, the dimensions of some of the elements in the figures may be exaggerated compared to other elements to help improve the understanding of embodiments of the present invention.

[0007] Where appropriate, the apparatus and process components have been represented by conventional symbols in the drawings, showing only those specific details essential for understanding the embodiments of the present invention, so as not to obscure the disclosure with details that are readily apparent to those skilled in the art who benefit from this description. Detailed description

[0008] In short, this provides an artificial intelligence (AI) server that offers natural language processing and query-response operation and is integrated into a communication system.

[0009] The AI ​​server, in its various embodiments, answers queries from one or more radios within the communication system. The different methods employed by the AI ​​server in these embodiments serve to minimize communication interference, maximize channel utilization, and prioritize responses. Operating according to one or more of these methods, the AI ​​server is advantageous for all types of communication systems, including, but not limited to, broadband systems, and even extends to broadband half-duplex and narrowband half-duplex systems, to name just a few.The advantages of extending the Kl server capabilities to half-duplex communication systems enable the public safety radio communication system to advantageously provide search query options via virtual search that were not readily available to public safety radio users in the past.

[0010] Fig.Figure 1A shows a communication system 100 comprising a wireless communication device 102 and an artificial intelligence (AI) server 104 operating in a communication network 106, according to some embodiments. The AI ​​server 104 includes speech processing and response capabilities to function as a virtual assistant. Virtual assistants, such as Siri, provided by Apple, Inc.®, and Google Now, provided by Google, Inc.®, are software applications that understand natural language and perform electronic tasks in response to user input. The communication device 102 can be any broadband or narrowband device with a microprocessor, a transceiver, and an audio circuit, such as a radio, a mobile phone, or the like, for receiving a spoken query user input for transmission to the AI ​​server 104.

[0011] The Kl-Server 104 manages query and response activity between the Kl-Server 104 and the communication device 102 in an optimized manner, thus preserving efficient use of the channel bandwidth. According to some embodiments, the Kl-Server 104 prioritizes responses sent to the query based on the verbosity of the response and the channel availability within the communication system 100. The verbosity response can be adapted in various ways to avoid exceeding the available channel bandwidth. For example, the Kl-Server can handle a verbosity response by prioritizing content within the verbosity response and subsequently segmenting the prioritized content into shorter, prioritized responses, thereby efficiently utilizing the available channel bandwidth without exceeding it.Content prioritization can also be based on factors such as the length and / or complexity of the detailed answer. Thus, various priority factors can be used to segment a response.

[0012] According to some embodiments, the Kl-Server 104 can also adjust the verbosity of the response by generating condensed responses. Adjusting verbosity through the use of a condensed response can be achieved in many ways, such as using acronyms instead of full words, using alternative shorter words, removing foreign words, and using predefined code names, to name a few. The Kl-Server can also take prioritized responses that have been previously defined and generate condensed responses. The completeness of the information in the condensed response can be adjusted based on channel availability.For example, shortening a response by removing less important information or sending less important information at a later time when channel availability has improved can improve overall operational efficiency. Therefore, the Kl-Server 104 embodiment enables improved operational efficiency of the communication system 100 by adjusting the verbosity of the response.

[0013] For additional efficient handling of channel bandwidth, the Kl-Server 104 can also provide an interrupt function for responses containing higher-priority content. Thus, if the communication device 102 is in the middle of reproducing information in response to a query relating to a general information facet, the Kl-Server can interrupt the response containing a higher-priority reply.

[0014] With reference to Fig.Method 1B is a method 150 with varying information for the optimized use of channel availability in a communication system according to some of the embodiments. Method 150 begins at 152 with receiving a query at an artificial intelligence (AI) server, such as AI server 104, wherein AI server has natural language processing and response capabilities. The query to AI server can be in the form of spoken queries or a combination of spoken and text queries.

[0015] Depending on the type of query sent, the communication server (KL) generates a verbose response to the query at 154. Moving to 156, the KL determines channel availability for the verbose response and then adjusts the verbosity of the response at 158 ​​based on the channel availability, which is based on channel bandwidth and channel occupancy. The response, adjusted to the query, is sent at 160. Therefore, communication system 100 and procedure 150 provide a KL server that determines the channel availability for a verbose response to a communication device and manages the verbosity of the response over available channels within the communication system.

[0016] In accordance with these embodiments of the method, adjusting the verbosity of the response can be achieved in several different ways, including, but not limited to: summarizing the content of the response to utilize channel bandwidth, and / or shortening the content of the response to avoid exceeding the available channel bandwidth. Adjusting the verbosity of the response can also be achieved by prioritizing the content of the detailed response and segmenting the prioritized content into shorter, segmented responses, thereby freeing up channel bandwidth between the responses.

[0017] Responses can be interrupted based on priority. For example, a customized response currently playing on a device can be interrupted by another response containing higher-priority content.

[0018] Therefore, the System 100 and the Method 150 embodiments offer the handling of a detailed response with which the available channel bandwidth can be used efficiently.

[0019] Fig.2 is a communication system 200, configured and functioning according to several embodiments. The communication system 200 comprises a plurality of communication devices 210 with a plurality of half-duplex radios, which may be portable handheld radios or mobile vehicle radios. These half-duplex radios communicate (transmit mode) using a push-to-talk (PTT) button over a communication channel with one or more of the other radios, which are listening (receive mode), and they are often referred to as two-way radios or PTT radios. While the Kl-Server 240 still provides the ability to vary a detailed response as described in the previous embodiments, it offers additional advantages relating to half-duplex communication systems.

[0020] The Communication System 200 can be a broadband system with PTT capability, such as that enabled over a wideband via PTT Server 280. The Communication System can also be a narrowband system, such as a public safety communication system used by law enforcement agencies, fire departments, and the like, encompassing the variety of portable and mobile PTT radios 210. Each of the radios 210 comprises a microprocessor, a transceiver, and suitable RF and controller circuitry for radio communication operation.

[0021] According to the embodiments, the artificial intelligence (AI) server 240 is integrated into the communication system 200 for answering queries from one or more of the half-duplex radios 210 that have formed a talkgroup 220. As already described, the AI ​​server is implemented with a natural language processing system and a spoken artificial query response system. Examples of such processing systems include, but are not limited to, Siri, OK Google, and other known or yet-to-be-developed systems.

[0022] During regular radio operation, talkgroup 220 can be assigned if any user in the group wishes to communicate with another user in the same talkgroup. A free radio channel is automatically found by the system (System 200), and the conversation takes place on that channel. Each radio transceiver, still controlled by its respective microprocessor, can participate in forming the talkgroup. Thus, creating talkgroup 220 allows radios within the multitude of radios (System 210) to be grouped together to listen for and respond to mutual communications on a separate, dedicated channel without involving the rest of the radio system.

[0023] According to the embodiments, the AI ​​server 240 utilizes talkgroup formation to respond to queries from members of the talkgroup 220. According to the embodiments, the AI ​​server intelligently interacts with a floor controller 250 to minimize communication interference, maximize channel utilization, and prioritize responses among talkgroup members. The integration of the AI ​​240 into the communication system 200 advantageously enables the half-duplex radios 220 to be operated as input points to receive verbal queries from a talkgroup member, thereby transforming radio operation into a virtual assistant.

[0024] The Kl-Server 240 in its various embodiments is capable of responding to requests and efficiently utilizing channel bandwidth. The Kl-Server 240 is advantageously suited for use in both narrowband and broadband communication systems with push-to-talk (PTT) capability.

[0025] In some embodiments, the Kl-Server 240 intelligently interacts with a Floor-Controller 250, which provides a variety of operational controls to minimize communication interference, maximize channel utilization, and prioritize responses between members of the talkgroup 220. For narrowband devices, such as land radios (LMRs), the Floor-Controller 250 can be located entirely within the Kl-Server 240 and / or operate as an independent floor control server. For applications extending the System 200 to broadband devices with PTT capability, such as certain PTT-enabled 3 / 4G LTE and Wi-Fi devices, the Floor-Controller 250 may be suitable for continued operation with and / or integration as part of a push-to-talk system on the Broadband Server 280.Servers and networks that support broadband PTT operation may, but are not limited to, include a WAVE™5000 server from Motorola Solutions, Inc.

[0026] According to the following embodiments, the system 200 offers the Kl-Server 240 for time-controlled query and response optimization and, in further embodiments, an additional query-to-query response function that enables access to supplementary information.

[0027] Initially, with reference to time-controlled query and response optimization, according to some embodiments, the Kl-Server 240 intelligently interacts with the Floor-Controller 250 to provide answers to queries sent by one or more radios from one or more talkgroups, and the timing of the answers from at least one of the Kl queries is prioritized based on context information from the communication system.

[0028] According to the embodiments, the AI ​​server 240 can respond to predefined verbal inputs or commands indicating an AI request from a talkgroup member to allow the AI ​​to join the talkgroup. A query sent from a radio in talkgroup 220 to the AI ​​server 240 can be optimized with respect to timing and prioritization. For example, a first talkgroup radio 216 initiates and sends a verbal request to the floor controller 250 to add artificial intelligence (AI) 240 within talkgroup 220. The AI ​​server 240 joins talkgroup 220 via 219.By adding the AI ​​to talkgroup 220, the AI ​​240 and the floor controller 250 are able to interact with the radios of talkgroup 220 to determine response times 262, identify a request or assign priority 264, adjust various floor controls 266, set delivery delays, delays based on confidence levels, and all other control functions related to timing and prioritization that can further improve the handling of the query-response system.

[0029] According to another query execution method, the assignment of the talkgroup to sub-talkgroups can also be changed based on the query sent by the originating radio. In response to a radio 218 sending a query to which the reply should only be played back to two specific radios 216 and 218, the KL server 240 responds by assigning the reply to sub-talkgroup (SUB-T / G) 222, so that the reply is only heard by members 216 and 218 of sub-talkgroup 222.

[0030] Accordingly, the system 200 enables the specification, for the Kl-Server 240, by the radio device sending the query, such as radio device 216, to a subgroup 222 of talkgroup members 216, 218 of talkgroup 220, who hear the answer, and the sending of the answer by the Kl-Server to the subgroup 222 of talkgroup members.

[0031] The Kl-Server 240 responds to further queries from members of the talkgroup 220 within the radio communication system 200. The Kl-Server interacts intelligently with a Floor-Controller 240 to minimize communication interference, maximize channel utilization, and prioritize responses between talkgroup members. Integrating artificial intelligence into a public safety radio communication system advantageously provides half-duplex radios with additional verbal and / or text query and response capabilities while maintaining regular talkgroup operation. The use of Kl is applicable to both narrowband and broadband communication systems with push-to-talk (PTT) capability.

[0032] Furthermore, in a public safety environment, it is important that the AI ​​server 240 provides useful information in response to user queries so as not to hinder time-critical public safety services (for example, responding to an emergency call, responding to an emergency in the correct location, and the like). Therefore, in some embodiments, it may be advantageous to alternatively assign the AI ​​server 240 automatically to each talkgroup when such talkgroups are formed within the communication system 200.

[0033] In some configurations, the Kl-Server 240 can insert itself into the conversation group based on specific keyword triggers. This self-insertion feature is particularly useful for announcements, events, and the like. For example, if a user of a portable device asks another member of the conversation group, "What time does Main Street close for the holiday parade?", the Kl-Server will insert itself into the conversation group and answer the query if no response is received from a member of the conversation group within a specified time period. For example: "Main Street will close between 10 a.m. and noon for the holiday parade." Even more advantageously, the Kl-Server 240 can provide additional information that an individual user might not be aware of, such as: "Detour planned at Second Street."

[0034] According to further embodiments, the Kl-Server 240 can automatically remove itself from the talkgroup if it cannot find a response to a query after a predetermined time, and even redirect the query, if available, to another source, such as the Dispatcher 230, Multimedia 270, or another member of the talkgroup. For example, the Multimedia resource 270 can provide a streaming video response to a redirected query. The Dispatcher 240 can be a narrowband or broadband dispatcher. This self-removal of the Kl advantageously preserves the efficiency of the system 200 by redirecting queries 249, 269 to other resources, thereby freeing the Kl-Server to handle other queries to which it can respond.

[0035] For embodiments where AI 240 redirects the query to another resource to obtain additional information, these queries can, for example, be redirected in text format to Dispatcher 230 to inquire whether the dispatcher has the resources to answer the query. The decision to send the query to Dispatcher 230 can be based on contextual factors associated with that dispatcher and on knowledge of the working environment of the various radio users of the multitude of radios 210 and the associated members working in talkgroups within the communication system 200. Floor control is automatically made available to Dispatcher 230 when it is confirmed that Dispatcher 230 does indeed have the information to answer the query. Dispatcher 230 then sends a verbal response via a dispatch radio to the query radio of talkgroup 220.

[0036] In other embodiments, the AI ​​240 can verbally query other radios in talkgroup 220 to determine if they possess information that can answer the query, in order to obtain additional information. If a radio member confirms such knowledge, the AI ​​server 240 can redirect the query to that radio member within talkgroup 220, also providing floor control to respond to the query. A response can then be issued by the initiating radio. Depending on the nature of the query sent, the response can be issued to all or some members of a talkgroup. If the query included a command or instruction to restrict the response to specific talkgroup members (radios 216, 218 of subgroup 222), the response can be restricted to those members as described previously.While it may be important for a response to reach one or more members, and for the group to be aware of this, it may not be necessary for the entire group to listen to the response. It may be sufficient to simply make certain members aware that a response has been given.

[0037] In another embodiment with additional information, the redirected query can be a text query sent to the multimedia resource 270, which can generate a streaming video response for transmission back to the initiating query radio 216. The streaming video can be output on all radios of sub-talkgroup 222, which in this example consists of radios 216 and 218. If no sub-talkgroup has been formed and no further restrictions have been imposed, the streaming video can be output via the radios of talkgroup 220.

[0038] In another embodiment with additional information, the redirected query can be a text query sent to the dispatcher 230, which operates in broadband mode. The dispatcher 230 can generate a video streaming response for transmission back to the initiating query radio 216. Therefore, a video streaming response to a verbal query originally triggered at the half-duplex radio 216 was provided by utilizing the query redirection capability of the Kl server.

[0039] According to further embodiments, the control of timing and the control of supplementary information are optimized by various floor control functions of the communication system 200. The floor controller 250 of the Kl server 230 provides floor control functions based on a variety of floor control factors 260, of which only a few are shown. Floor control functions can be based on the length of an expected response. For example, the Kl server 240 can request and / or lock the floor for a predetermined time to complete a response 262.

[0040] The priority of a Kl-Floor request may depend on the priority of the requesting radio. Some radio users, identified by their radio user ID, and / or some types of verbal requests may have a higher priority and be answered before others.

[0041] When a channel is heavily used, the KL server can adjust the content and / or depth of a response to fill the available floor time. For example, a heavily used radio channel can have a response converted into a summary response to avoid impacting channel usage.

[0042] In some embodiments, it may be desirable to prioritize the order of the response content and segment it into smaller responses in order to free up channel availability between responses.

[0043] The priority (and therefore the ability to interrupt the AI) of other members of the conversation group can be adjusted based on the applicability of the content to them. For example, if a member of the conversation group is listening to a response and determines that the response is no longer relevant to the current conditions of the conversation group or requires further details, this higher-priority user can interrupt the response by pressing PTT and formulating a new query. For instance, if a rookie police office sent a request asking to stream a full map of the entire holiday parade route via Multimedia 270 219, then a senior office could interrupt this response with a query: "Does the holiday parade cross Main Street and Second Street?"

[0044] Accordingly, communication systems, such as a public safety communication system, and half-duplex radios operated within the system, can advantageously enable the operation of talkgroups with additional query search capabilities through the integration of floor control functions with artificial intelligence.

[0045] Fig. Figure 3 shows an example of a communication exchange diagram in which artificial intelligence is added as a member of the talkgroup, according to some of the embodiments. At 310, a first sub-talkgroup radio 301 sends a verbal request to add artificial intelligence (AI) within the talkgroup. This request is answered by a group manager 350 (corresponding to the floor controller in Figure 3). Fig.2) Received at 310. The request 310 triggers a response from the group manager 350, and at 312, artificial intelligence 340 (AI) is added to the talkgroup. By adding AI 340 to the group, AI 340 and the floor manager 350 are able to interact with the talkgroup's radios to determine response times, increase or decrease delivery, set priorities, assign priorities, and all other control functions that can further improve the handling of the query response system.

[0046] A query is sent from a second subgroup radio, 304, to the AI ​​340 at station 314. AI 340 performs word analysis to determine a confidence level that the query is intended for the AI. Depending on the confidence level, a request for a delay is sent to the floor, 350, as specified at station 316. AI 340 then sends a request to the group manager, 350, for floor time at a specific time T and for a predetermined duration Td. The group manager, 350, then grants the floor to AI 340 at station 318. AI 340 has already completed the search and is ready at station 320 to generate and send a response to radios 302 and 301. After this, the floor can be removed at station 322.

[0047] The AI's response is sent to the query radio 302 and automatically also to the initiating radio 301, unless the requester specifies otherwise. These radios all operate on the same sub-talkgroup channel, so unless there is an instruction to assign or redirect the response to another free channel, all radios within the talkgroup will hear the response.

[0048] Thus, the illustrated time axis 300 has clarified some of the timing factors that can be taken into account, according to the embodiments.

[0049] With reference to Fig. Figure 4 describes a method 400 according to an embodiment with query-to-query. The query-to-query embodiment makes it possible to obtain supplementary information beyond that normally provided by the Kl-Server 240. Fig.Beyond the 2 available. Starting at 402, a talkgroup is formed from a multitude of radios, followed by triggering a query on one of the radios via the Push-To-Talk (PTT) button at 404.

[0050] The procedure can further include triggering the AI ​​server via a radio input prior to initiating the query, allowing the AI ​​server to join the talkgroup. The radio input (the trigger) could be, for example, a verbal PTT pre-command and / or a non-PTT out-of-band text message, depending on the system. The search query can follow the pre-command. For example, the pre-command could be the spoken word "Einstein," followed by the search query "How many registered firearms are located at this location?"

[0051] Procedure 400 continues at 406 by transferring the search query to the AI ​​server for artificial intelligence, where the AI ​​server is 240 of the AI ​​server. Fig. 2 is, which, as previously described, exhibits natural language processing and response capabilities.

[0052] At 408, according to the query-to-query implementation, the AI ​​server determines, in response to receiving the query, that an alternative response resource is capable of answering the query or providing a better answer. The AI ​​server 240 then requests the floor controller 250 at 410 to transfer the talkgroup floor to the alternative response resource, such as the dispatcher 230 or another radio in the talkgroup. Fig.2. The verbal query is redirected at 412 from AI Server 240 via the Floor Controller to the alternative resource. In some embodiments, it may be useful for the AI ​​Server to convert the verbal query into a text query before forwarding it to the alternative resource. For example, verbal queries converted to text format can be redirected from AI Server 240 to Dispatcher 230. The text format is preferred to avoid interfering with the dispatch of audio radio communications that can continue with other users. One advantage of redirecting a query request to Dispatcher 230 is that this dispatch resource can research the text version of the query and return a verbally researched response to the initiating query radio via the dispatch transmission.The response is thus provided directly by the dispatcher 230 to the triggering query radio 216, without having to go back via the AI ​​server 240, which further increases the efficiency of the operation.

[0053] If the KL server does not know the answer to a query or determines that an alternative source would have a better answer, the KL server can also redirect the query to that source. For example, another member of the conversation group or a multimedia source 270.

[0054] The Floor Controller 250 automatically makes the floor available to the alternative resource when the Kl Server 240 requests additional information for the query or redirects the query. Using alternative resources offers improved utilization of channel bandwidth and efficiency in handling the query-response portion of the communication system.

[0055] Continue to Fig.5, a method 500 is presented which the in Fig. The method 500 summarizes the timing implementation forms described in the two. Method 500 begins at 502 with the formation of a talkgroup, followed by the transmission of a query from a PTT radio at 504. In some embodiments, the Kl server can automatically insert itself and later remove itself from the talkgroup after sending the reply. In some embodiments, the Kl server 240 and a floor controller 250 can be automatically assigned when one or more talkgroups are formed within a communication system. In still other embodiments, a radio member of the talkgroup sends a Kl request, as described in the two ... Fig. As described in section 3, to join the KL server and the discussion group.

[0056] According to some embodiments, controlling the timing and prioritization of the response sent from the Kl-Server to the PTT device will significantly improve the overall handling of the query response system. For example, the Kl-Server 240 offers... Fig.2. The system waits until a predetermined response time has elapsed before generating and sending the response, giving members of the conversation group the opportunity to provide relevant information and alert other members. Using the predefined response time ensures that a response to the query can still be received from the KL server even if no member of the conversation group is able to respond. The response time can be determined based on query analysis to obtain a confidence level that the query was intended for the KL 240, the radio channel bandwidth, and the availability of the KL server's floor control, or any combination thereof, enabling efficient channel utilization control. Analyzing the query to determine the confidence level may involve a name search.For example, if the AI ​​is named Einstein and the query includes the name "Einstein," there is a very high probability that it is a query directed to AI 240. With such high confidence, AI 240 can respond immediately, resulting in zero waiting time. However, if the query begins with the name of another user in the conversation group, it is likely that the query was not directed to the AI ​​server. In this case, the AI ​​waits the maximum waiting time before responding. If, during the waiting period, the AI ​​detects that the queried user has responded, AI 240 will not respond unless it has additional information that might be useful.

[0057] Procedure 500 can be further enhanced, if necessary, by applying priorities. For example, by assigning a priority to sending a response from AI Server 240 and sending the response based on that priority. Prioritizing a radio query and its subsequent response can be based on contextual factors, such as the identity of the radio user, the rank of the requester, the rank of other members of the talk group, and the number of members in the talk group. For example, a fire chief, a police officer, or a detective, to name a few. Prioritizing the radio query can also be based on contextual factors related to the location of the radio query, such as the type of location and public safety information concerning that location.Examples include traffic accidents, airplane explosions, train derailments, robberies, and trespassing. Radio interrogation prioritization can be based on verbal query words with predefined keyword priorities such as "FIRE," "TOXIC," "EMERGENCY," "POISON," and "EXPLOSION," among others.

[0058] According to some embodiments, the method 500 can further adjust a response waiting time by the Kl-Server based on a confidence level that the verbal query was intended for the Kl-Server or for members of the discussion group.

[0059] According to some embodiments, the method 500 can further determine an expected response time. Depending on the type of priority, it may also be desirable to lock the floor control to the expected response time. For example, in mission-critical events, such as fire rescues, generating and transmitting a high-priority response is critical when other members of a conversation group have been unable to answer a query, and there is a fairly high level of confidence that the intended recipient of the query is now the KL server 240.

[0060] Procedure 500 can be further improved by re-prioritizing automated verbal responses to the query by the AI ​​server 240 in response to changes in the AI ​​server-determined context of the event location, with the context of the event location being monitored as part of the query.

[0061] The following table contains some examples of confidence levels, channel allocation, KL priority, waiting time and required response time. Type of applicant query Confidence level that the query is suitable for AI (High / Medium / Low) Channel availability (High / Medium / Low) Class priority (High / Medium / Low) Waiting time Required response length - time (seconds) Effect timeline / Verbosity / Query-to-Query information Fire Chief Einstein, what kind of flammable materials are in this building? H N H 0 30 Timeline AI requests channel availability for 30 seconds with high priority to send the response "Chemicals x,y,z". policeman Who are the members of the ABC gang? M N M 10 10 Timeline AI waits 10 seconds to see if anyone responds; if not, it requests the floor for 5 seconds to see the response "Member names Lee, Barb". detective What license plates are on the red corvette? N N M 10 10 Query-Query-Information & Timeline: AI has no answer, but determines that Officer Smith should know the answer. AI waits for the officer to respond; if no response is received, it requests floor control. "Officer Smith, please provide license plate numbers from your position." AI transfers floor control to Officer Smith.

[0062] The table is intended to provide non-restrictive examples of a few scenarios in which the different implementations are applied.

[0063] Accordingly, a communication system was provided that incorporates artificial intelligence and methods for controlling an AI server within the system. A method for varying the verbosity of the response, a method for determining the timing of a response, and a method for obtaining supplementary information for a response were provided. The system and methods have enabled optimized use of channel bandwidth, improved timing, and query / response redirection for more accurate information gathering. Improved handling of search queries and virtual assistant responses can now be achieved using an artificial intelligence (AI) server, which is managed according to the described methods of the various implementations.

[0064] Specific embodiments have been described in the preceding specification. However, it is clear to those skilled in the field that various modifications and changes can be made without departing from the spirit of the invention, as set out in the claims below. Accordingly, the specification and the figures are to be understood in an illustrative rather than a restrictive sense, and all such modifications are to be included within the scope of protection of the present teachings.

[0065] The benefits, advantages, problem solutions, and any conceivable element that leads to or enhances any benefit, advantage, or solution shall not be construed as critical, necessary, or essential features or elements of any claim or all claims. The invention is defined exclusively by the attached claims, including any amendment made during the pendency of the present application and all equivalents of such claims as published.

[0066] Furthermore, in this document, relational expressions such as first and second, above and below, and the like are to be used solely to distinguish one entity or action from another, without necessarily requiring or implying any actual relationship or order between such entities or actions. The expressions "includes," "comprising," "has," "having," "include," "containing," "containing," or any variation thereof are to cover non-exclusive inclusion, so that a process, procedure, article, or device that includes, has, includes, or contains a list of elements may not only include such elements but may also include other elements not expressly listed or inherent in such processes, procedures, articles, or devices. An element that continues with "includes... a," "has..."The terms "one," "includes... one," and "contains... one" do not, without further stipulations, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprise, have, include, or contain the element. The terms "one" and "a" are defined as one or more unless explicitly stated otherwise herein. The terms "essentially," "essentially," "approximately," "about," or any other version thereof are defined as "being close to" as is clear to those skilled in the art, and in one non-limiting embodiment, the term is defined as being within 10%, in another embodiment within 5%, in another embodiment within 1%, and in yet another embodiment within 0.5%. The term "coupled," as used herein, is defined as "connected," although not necessarily directly and not necessarily mechanically.A device or structure that is “configured” in a certain way is configured at least in that way, but may also be configured in at least one other way not listed.

[0067] It is desired that some embodiments include one or more generic or specialized processors (or “processing devices”), such as microprocessors, digital signal processors, custom processors and freely programmable field-gate arrays (FPGAs) and unique stored program instructions (comprising both software and firmware) that control the one or more processors to implement, in conjunction with certain non-processor circuitry, some, most or all of the functions of the method and / or device described herein.Alternatively, some or all functions can be implemented by a state machine that has no stored program instructions, or in one or more application-specific integrated circuits (ASICs) where each function, or some combinations of certain functions, are implemented as custom logic. Naturally, a combination of the two approaches can be used.

[0068] Furthermore, an embodiment can be implemented as a computer-readable storage medium containing computer-readable code stored thereon for programming a computer (which, for example, includes a processor) to perform a method described and claimed herein. Examples of such computer-readable storage media include, but are not limited to: a hard disk, a CD-ROM, an optical storage device, a magnetic storage device, a ROM (read-only memory), a PROM (programmable read memory), an EPROM (erasable programmable read memory), an EEPROM (electrically erasable programmable read memory), and flash memory.Furthermore, it can be expected that a person skilled in the art, regardless of possible considerable effort and a large selection of designs, which is justified, for example, by available time, current technology and economic considerations, guided by the concepts and principles disclosed herein, will be able to produce such software instructions, programs and ICs with minimal experimental effort.

[0069] The summary of the disclosure is provided to allow the reader to quickly grasp the nature of the technical disclosure. It is submitted with the understanding that it is not intended to interpret or limit the spirit or meaning of the claims. Furthermore, it is clear from the preceding detailed description that various features in different embodiments are grouped together to streamline the disclosure. This method of disclosure should not be interpreted as reflecting an intention that the claimed embodiments require more features than are expressly stated in each claim. Rather, as is evident from the following claims, an inventive subject matter is present in fewer than all the features of any single disclosed embodiment.Thus, the following claims are integrated into the detailed description, with each claim standing alone as a separately claimed subject matter.

Claims

[1] Method (400) for receiving information, comprising: forming a conversation group (220) from a large number of radios (210); Sending a query via a (push-to-talk) PTT radio of the talkgroup (220) to at least one member of the talkgroup (220); and Generating and sending a response to a query from a KL server (240) with natural language processing and response capability after a predetermined response waiting time for the conversation group (220) has elapsed without a response being generated by the conversation group (220). [2] The method of claim 1, further comprising: Receiving a response to a query from another radio (210) of the talkgroup (220) before the waiting time has elapsed. [3] Method according to claim 1, further comprising: Receiving a response to the query from the KL server (240) after the specified waiting time has elapsed. [4] Method according to claim 1, wherein the waiting time is determined based on a radio channel bandwidth and the availability of a Kl server floor control (250). [5] The method of claim 1 further comprising: Determining a priority (264) for sending a response from the Kl server (240); and sending the response based on the priority (264). [6] The method of claim 1, further comprising: Prioritizing radio queries by the Kl server (240) or a floor controller (250), based on context factors. [7] Method according to claim 6, wherein the context factors are associated with the discussion group (220) and comprise at least one of: Identity of a radio user, rank of the requester, rank of other members of the talkgroup (220), number of members in the talkgroup (220). [8] Method according to claim 6, wherein the context factors are associated with an event location and comprise: a type of incident location and public safety information concerning the incident location. [9] Method according to claim 1, further comprising: Prioritizing a radio query based on verbal query words with predetermined keyword priorities. [10] The method of claim 9, further comprising: The reprioritization of automated verbal responses to the query by the Kl-Server (240) in response to changes in the event location context determined by the Kl-Server (240), where the event location context is monitored as part of the query. [11] Method according to claim 1, further comprising: the automatic assignment of the KL server (240) and a floor controller (250) when forming one or more talk groups (220) within a communication system (200). [12] Method according to claim 1, further comprising: adjusting a response waiting time by the Kl-Server (240) based on a confidence level that the verbal query was intended for the Kl-Server (240) or for members of the conversation group (220). [13] Method according to claim 1, further comprising: Determining an estimated response time (262); locking the floor control (266) for the expected response time (262). [14] The method of claim 1, further comprising: the specification, for the KL server (240), by the radio (210) that sends the request, to a subset (222) of talkgroup members who hear the response; and the sending of the response by the KL server (240) to the subgroup (222) of discussion group members. [15] Method according to claim 1, further comprising: where the KL server (240) automatically removes itself from the conversation group (220) after sending the reply. [16] Communication system (200), comprising: a plurality of radio devices (210) with half-duplex functionality, wherein the plurality of radio devices (210) forms one or more talkgroups (220); a server (240) for artificial intelligence (Kl) that provides a query and response database with natural language processing; a floor controller (250) for planning the query and response process; wherein the Kl server (240) intelligently interacts with the floor controller (250) to provide answers to queries sent by one or more radios (210) from one or more talkgroups (220) after a predetermined response waiting time for the talkgroup (220) has elapsed without a response being generated by the talkgroup (220); and where the timing of responses to at least one of the AI ​​queries is prioritized based on context information from the communication system. [17] Communication system according to claim 16, wherein the talk group comprises half-duplex radios operating in a narrowband network. [18] Communication system according to claim 16, wherein the talk group comprises half-duplex radios operating in a broadband network. [19] Communication system according to claim 16, wherein the talk group is a public safety talk group and the KL server and the floor controller prioritize the timing of responses to requests based on context information within the public safety communication system. [20] Communication system (200), comprehensive, a push-to-talk (PTT) communication device with group talk capability; a server (240) for artificial intelligence (AI) with speech processing and response capabilities; and a floor controller (250) for interoperative control of timing and prioritization of query and response between the PTT device (210) and the Kl server (240), wherein the Kl server (240) generates and sends a response to a query from the PTT communication device (210) to the talk group (220) after a predetermined response waiting time for the talk group (220) has elapsed without a response being generated by the talk group (220). [21] Communication system according to claim 20, wherein the talk group (220) comprises half-duplex radio equipment operated in a narrowband network. [22] Communication system according to claim 20, wherein the talk group comprises half-duplex radio equipment (210) operated in a broadband network. [23] Communication system according to claim 20, comprising, wherein the Kl server (240) automatically inserts itself into the talk group (220) without being queried and sends a response to a query radio of the talk group (220).

Images