WO2020233075A1 - 基于语音识别的会议室预定方法、装置、设备及存储介质 - Google Patents

基于语音识别的会议室预定方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2020233075A1
WO2020233075A1 PCT/CN2019/122064 CN2019122064W WO2020233075A1 WO 2020233075 A1 WO2020233075 A1 WO 2020233075A1 CN 2019122064 W CN2019122064 W CN 2019122064W WO 2020233075 A1 WO2020233075 A1 WO 2020233075A1
Authority
WO
WIPO (PCT)
Prior art keywords
room
parameters
conference room
meeting
conference
Prior art date
Application number
PCT/CN2019/122064
Other languages
English (en)
French (fr)
Inventor
王金燕
Original Assignee
深圳壹账通智能科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳壹账通智能科技有限公司 filed Critical 深圳壹账通智能科技有限公司
Publication of WO2020233075A1 publication Critical patent/WO2020233075A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/02—Reservations, e.g. for tickets, services or events
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue

Definitions

  • This application relates to the field of artificial intelligence technology, and in particular to a method, device, equipment, and storage medium for booking a meeting room based on voice recognition.
  • the present application provides a method, device, equipment, and storage medium for booking a meeting room based on voice recognition, aiming to simplify the process of booking a meeting room and improve the intelligence of meeting room booking.
  • this application provides a method for booking a meeting room based on voice recognition, the method including:
  • the step of recognizing the reserved voice of the conference room and obtaining the parameters of the conference room to be reserved includes:
  • Preprocessing the predetermined voice of the conference room cut off the mute at the beginning and the end of the predetermined voice of the conference room, and divide the predetermined voice of the conference room into frames to obtain multiple frames;
  • a plurality of the keywords are respectively filled in a preset meeting room parameter template to be booked to obtain the meeting room parameter to be booked.
  • the present application also provides a voice recognition-based conference room reservation device.
  • the voice recognition-based conference room reservation device includes:
  • the receiving module is used to receive the conference room reservation voice sent by the user
  • the voice recognition module is used to recognize the reserved voice of the conference room and obtain the parameters of the conference room to be reserved;
  • a calculation module configured to calculate the matching degree between the parameters of the conference room to be booked and the parameters of the available conference room based on a preset matching degree calculation formula
  • a recommendation module configured to display the corresponding available meeting rooms whose matching degree is higher than the preset matching degree threshold on the meeting room recommendation interface for users to choose;
  • the voice recognition module is also used for:
  • Preprocessing the predetermined voice of the conference room cut off the mute at the beginning and the end of the predetermined voice of the conference room, and divide the predetermined voice of the conference room into frames to obtain multiple frames;
  • a plurality of the keywords are respectively filled in a preset meeting room parameter template to be booked to obtain the meeting room parameter to be booked.
  • the voice recognition-based conference room reservation device includes a processor, a memory, and a voice recognition-based conference room reservation computer stored in the memory.
  • a readable instruction when the computer readable instruction for booking a meeting room based on voice recognition is executed by the processor, the steps of the method for booking a meeting room based on voice recognition as described above are implemented.
  • this application also provides a computer storage medium, the computer storage medium stores a computer readable instruction for booking a meeting room based on voice recognition, and the computer readable instruction for booking a meeting room based on voice recognition is processed by the processor The steps of the method for booking a meeting room based on voice recognition as described above are implemented at runtime.
  • the present application proposes a method, device, equipment, and storage medium for a meeting room reservation based on voice recognition.
  • the method includes: receiving a meeting room reservation voice sent by a user; and recognizing the meeting room reservation voice to obtain Meeting room parameters to be booked; calculating the matching degree between the meeting room parameters to be booked and the available meeting room parameters based on the preset matching degree calculation formula; displaying the corresponding available meeting rooms whose matching degree is higher than the preset matching degree threshold Recommend the interface in the meeting room for users to choose.
  • this application obtains the meeting room parameters to be booked based on speech recognition, and then obtains the matching degree between the meeting room parameters to be booked and the available meeting room parameters according to the matching degree calculation formula, and the matching degree is higher than the preset
  • the available meeting rooms that match the threshold are recommended to users, thereby simplifying the meeting room reservation process and improving the intelligence of meeting room reservations.
  • FIG. 1 is a schematic diagram of the hardware structure of a conference room reservation device based on voice recognition according to various embodiments of the present application;
  • FIG. 2 is a schematic flowchart of a first embodiment of a method for booking a meeting room based on speech recognition in this application;
  • FIG. 3 is a schematic flowchart of a second embodiment of a method for booking a meeting room based on voice recognition in this application;
  • Fig. 4 is a schematic diagram of functional modules of a first embodiment of a device for booking a meeting room based on voice recognition in this application.
  • the voice recognition-based conference room reservation device mainly involved in the embodiments of the present application refers to a network connection device capable of realizing network connection.
  • the voice recognition-based conference room reservation device may be a server, a cloud platform, and the like.
  • FIG. 1 is a schematic diagram of the hardware structure of a speech recognition-based conference room reservation device related to various embodiments of the present application.
  • the conference room reservation device based on voice recognition may include a processor 1001 (for example, a central processing unit, a CPU), a communication bus 1002, an input port 1003, an output port 1004, and a memory 1005.
  • the communication bus 1002 is used to realize the connection and communication between these components; the input port 1003 is used for data input; the output port 1004 is used for data output.
  • the memory 1005 can be a high-speed RAM memory or a stable memory (non-volatile memory). memory), such as a disk memory.
  • the memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.
  • the memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.
  • the memory 1005 as a readable storage medium in FIG. 1 may include an operating system, a network communication module, an application program module, and computer-readable instructions for booking a meeting room based on voice recognition.
  • the network communication module is mainly used to connect to the server and perform data communication with the server; and the processor 1001 can call the voice recognition-based conference room reservation computer-readable instructions stored in the memory 1005, and execute the instructions provided in the embodiments of this application. The method of booking a meeting room based on voice recognition.
  • the embodiment of the present application provides a method for booking a conference room based on voice recognition.
  • Fig. 2 is a schematic flowchart of a first embodiment of a method for booking a meeting room based on voice recognition in this application.
  • the voice recognition-based conference room reservation method is applied to a voice recognition-based conference room reservation device, and the method includes:
  • Step S101 receiving a conference room reservation voice sent by a user
  • conference room reservations require manual recording of the reservation situation. Even if some companies have developed online conference room reservation products, users still need to enter the required conference room in the preset conference room configuration interface during the conference room reservation process Detailed information, such as meeting room number, occupancy time, etc.
  • the recording function of the meeting room reservation device based on voice recognition is used to receive the user’s meeting room reservation voice.
  • the meeting room reservation voice includes the parameters of the meeting room to be reserved, and the parameters of the meeting room to be reserved include the floor and the starting point. One or more of the start time, the duration of the meeting, and the number of participants.
  • the method further includes:
  • the voice input prompt is displayed on the display screen of the conference room reservation device based on voice recognition, so that the user can refer to the voice input prompt for voice input.
  • the voice input prompt is displayed on the display screen of the conference room reservation device based on voice recognition, or a pre-recorded voice input prompt file is played by voice.
  • the voice input prompt may be: I need to book a meeting room for X people in Building XX from X o'clock to Y o'clock this afternoon.
  • Step S102 Recognizing the reservation voice of the conference room, and obtaining parameters of the conference room to be reserved;
  • the step of recognizing the reserved voice of the conference room and obtaining the parameters of the conference room to be reserved includes:
  • Step S102-1 preprocessing the scheduled voice of the conference room, cut off the mute at the beginning and end of the scheduled voice of the conference room, and divide the scheduled voice of the conference room into frames to obtain multiple frames;
  • the mute includes mute at the beginning and end of the scheduled voice of the conference room, thereby eliminating the mute to save resources and reduce interference without reducing service quality. Further, cut the predetermined voice of the conference room with the silence cut into small segments, such as one small segment every 25ms, and each small segment is called a frame, so that multiple frames can be obtained after the frame is divided. Generally, It can be achieved by using a moving window function, and there are generally overlapping parts between adjacent frames. If the length of each frame is 25ms, an overlap of 15ms can be set between adjacent frames. In this way, the frame length is 25ms and the frame shifts 10ms for framing.
  • Step S102-2 extracting feature vectors in multiple frames, and obtaining text corresponding to the feature vectors
  • the scheduled speech in the conference room becomes multiple frames, but the waveform does not have the ability to describe in the time domain, so further waveform transformation must be performed.
  • the multiple frames are windowed to obtain multiple short-term analysis windows to obtain the corresponding frequency spectrum; and the filter is used to filter the frequency spectrum to obtain the filtered frequency spectrum; the filtered frequency spectrum is obtained by taking the logarithm and inverse Cepstrum analysis is performed in a transformed manner to obtain the feature vector of the predetermined voice of the conference room.
  • each frame in the predetermined voice of the conference room is converted into a multi-dimensional feature vector according to the acoustic feature in the predetermined voice of the conference room, and the feature vector includes content information of the predetermined voice of the conference room in the frame.
  • the step of obtaining the text corresponding to the feature vector includes:
  • the acoustic model may output phoneme information corresponding to the feature vector.
  • the language model is obtained by training a large amount of text information, and the probability that a single character or word is related to each other can be obtained by using the language model.
  • Step S102-3 Obtain multiple keywords in the text
  • keywords related to the reservation of the conference room are acquired, and redundant text is removed.
  • a keyword library is preset, and the keyword library includes keywords related to conference room reservations, such as numbers, time, and floors. Mark the words and ⁇ or words in the text that are the same as those in the keyword library as keywords. Understandably, there are multiple keywords.
  • step S102-4 a plurality of the keywords are respectively filled in a preset meeting room parameter template to be booked to obtain the meeting room parameter to be booked.
  • a parameter template of the meeting room to be booked is preset, and the parameter template of the meeting room to be booked includes a plurality of filling positions of the keywords, so as to fill in the keywords to obtain the meeting room to be booked parameter.
  • the parameter template of the conference room to be booked can be set as: the floor is (X F ), the start time is (X S ), the duration is (X L ), and the number of participants is (X N ); the X in the brackets F , X S , X L and X N are the numerical values of the corresponding content.
  • the method further includes:
  • Step S102a Extract voiceprint information in the predetermined voice of the conference room, and compare the voiceprint information with pre-stored voiceprint information;
  • one or more predetermined voiceprint information is collected in advance, the meeting room reservation authority is configured for the predetermined voiceprint information, and detailed information of the user corresponding to the predetermined voiceprint information of the meeting room is recorded, such as name, contact number , Mailbox, position, etc.
  • the meeting room reservation authority is configured for the predetermined voiceprint information
  • detailed information of the user corresponding to the predetermined voiceprint information of the meeting room is recorded, such as name, contact number , Mailbox, position, etc.
  • the voiceprint information is consistent with the pre-stored voiceprint information, it is determined that the user corresponding to the voiceprint information has the predetermined weight of the meeting room, and then the step is performed: identifying the predetermined voice of the meeting room, and obtaining the parameters of the meeting room to be reserved; If the voiceprint information is inconsistent with the pre-stored voiceprint information, it is determined that the user corresponding to the voiceprint information does not have the weight of the conference room reservation, and the conference room reservation voice is ignored. As a result, users without authority can be prevented from making conference room reservations, which facilitates the management of conference room reservations.
  • Step S103 Calculate the matching degree between the parameters of the conference room to be booked and the parameters of the available conference room based on a preset matching degree calculation formula
  • the step before the step of calculating the matching degree between the to-be-booked meeting room parameters and the available meeting room parameters based on the preset matching degree algorithm, the step further includes: pre-saving the meeting room parameters and usage status of all meeting rooms;
  • the meeting room whose usage status is free in a specific time period is marked as an available meeting room, and the meeting room parameters of the available meeting room are marked as the available meeting room parameters, wherein the specific time period is pre-defined
  • the set time floating range and the starting time and duration in the predetermined parameters of the meeting room are determined.
  • the usage status includes in-use, idle, and maintenance status.
  • the usage status is related to time, so the meeting room that is free in a specific time period is marked as an available meeting room.
  • the specific time period is determined by a preset time floating range and the start time and duration in the predetermined parameters of the conference room.
  • the time floating range can be specifically set, for example, a certain period of time before and after the start time, such as 60min, 120min, etc.; or the start time point moves forward or backward by a preset number of days, such as the start time point One day forward or several days forward.
  • the step of calculating the matching degree between the parameters of the conference room to be booked and the parameters of the available conference room based on the preset matching degree calculation formula includes:
  • the parameters of the conference room to be booked and the parameters of the available conference room may include more or less parameters than F, S, L, and N.
  • Step c express the matching degree calculation formula as:
  • C represents the matching degree
  • ⁇ F , ⁇ S , ⁇ L , ⁇ N respectively represent the floor weighting factor, the starting time weighting factor, the meeting duration weighting factor, and the number of people weighting factors, ⁇ F , ⁇ S , ⁇ L , ⁇
  • the value of N is respectively preset according to the floor difference, the start time difference, the meeting duration difference, and the number of people difference in the parameters of the meeting room to be reserved and the parameters of the available meeting room. Understandably, when a certain available conference room parameter is completely consistent with the to-be-booked conference room parameter, the matching degree C is 100; when a certain available conference room parameter is more different from the to-be-booked conference room parameter , The smaller the matching degree C is.
  • x 1 , x 2 , x 3 , x 4 , and x 5 represent the floor difference.
  • x 1 ⁇ x 2 ⁇ x 3 ⁇ x 4 ⁇ x 5 are all positive integers.
  • the difference determines the value of the floor weighting factor ⁇ F : when the floor difference is between [0, x 1 ], the value of ⁇ F is ⁇ F1 ; when the floor difference is between [x 2 , x 3 ], the value of ⁇ F is ⁇ F2 ; when the floor difference is between [x 4 , x 5 ], the value of ⁇ F is ⁇ F3 ; where ⁇ F1 , ⁇ F2 ,
  • the value of ⁇ F3 can be set according to experience, for example, set to 0.8, 1.0 and 1.2 respectively.
  • y 1 , y 2 , y 3 , y 4 , y 5 represent the start time difference
  • y 1 ⁇ y 2 ⁇ y 3 ⁇ y 4 ⁇ y 5 and the start time difference is determined according to the start time difference.
  • the value of the time weighting factor ⁇ S when the floor difference is between [0, y 1 ], the value of ⁇ S is ⁇ S1 ; when the floor difference is between [y 2 , y 3 ] , The value of ⁇ S is ⁇ S2 ; when the floor difference is between [y 4 , y 5 ], the value of ⁇ S is ⁇ S3 ; where ⁇ S1 , ⁇ S2 , ⁇ S3 are The value can be set according to experience, such as 0.8, 1.0 and 1.2 respectively.
  • z 1 , z 2 , z 3 , z 4 , z 5 represent the difference in duration of the meeting
  • z 1 ⁇ z 2 ⁇ z 3 ⁇ z 4 ⁇ z 5 and the difference in duration of the meeting is determined session duration factor ⁇ L values: when the difference in duration of the session between [0, 1 z], ⁇ L is the value of ⁇ L1; when the difference in duration of the session [z 2, z 3 ], the value of ⁇ L is ⁇ L2 ; when the difference in duration of the meeting is between [z 4 , z 5 ], the value of ⁇ L is ⁇ L3 ; where ⁇ L1
  • the values of ⁇ L2 and ⁇ L3 can be specifically set as required, for example, set to 0.8, 1.0 and 1.2 respectively.
  • m 1 , m 2 , m 3 , m 4 , m 5 represent the difference in the number of people
  • m 1 , m 2 , m 3 , m 4 , and m 5 are all positive integers
  • m 1 ⁇ m 2 ⁇ m 3 ⁇ m 4 ⁇ m 5 determining a value of the difference between the number of persons ⁇ N weighting factors according to: when the difference in the number of [0, m 1] between the values of ⁇ N ⁇ N1, when the difference in the number [m 2, m 3] between the value of ⁇ N is ⁇ N2, when the difference in the number [m 4, m 5] between the value of ⁇ N Is ⁇ N3 ; where the values of ⁇ N1 , ⁇ N2 , and ⁇ N3 can be specifically set as required, for example, set to 0.8, 1.0 and 1.2 respectively.
  • Step d Substitute the weighting factor, the available meeting room parameters, and the to-be-booked meeting room parameters into the matching degree calculation formula to calculate the matching degree.
  • the corresponding ⁇ F , ⁇ S , ⁇ L , and ⁇ N are obtained according to the parameters of the meeting room to be booked and the available meeting room parameters; and the available meeting room parameters,
  • the degree of matching between the parameters of the meeting room to be reserved it is understandable that the degree of matching, the parameters of the available meeting room and the parameters of the meeting room to be booked correspond to each available meeting room and the meeting room to be booked. Since there are one or more available conference rooms, the matching degree also corresponds to one or more.
  • Step S104 Display the corresponding available meeting rooms with the matching degree higher than the preset matching degree threshold on the meeting room recommendation interface for the user to select.
  • the matching degree threshold is preset, for example, 80%. If a certain available conference room parameter is completely consistent with the to-be-booked conference room parameter, the matching degree C is 100; if the difference between a certain available conference room parameter and the to-be-booked conference room parameter is smaller, the matching The greater the degree C is; if the difference between a certain available conference room parameter and the conference room parameter to be reserved is greater, the matching degree C is smaller. Therefore, in this embodiment, the corresponding available meeting room with the matching degree higher than the preset matching degree threshold is displayed on the meeting room recommendation interface for the user to select.
  • a queuing function can be set to receive a queuing instruction sent by a user, and when there is an available meeting room with a higher matching degree among the available meeting rooms, the meeting room with a higher matching degree is recommended to the user.
  • the conference room reservation voice sent by the user is received; the conference room reservation voice is recognized to obtain the conference room parameters to be reserved; the parameters of the conference room to be reserved and the available conference rooms are calculated based on the preset matching degree calculation formula
  • the matching degree between the parameters; the corresponding available meeting room whose matching degree is higher than the preset matching degree threshold is displayed on the meeting room recommendation interface for the user to select.
  • This application obtains the meeting room parameters to be booked based on speech recognition, and then obtains the matching degree between the meeting room parameters to be booked and the available meeting room parameters according to the matching degree calculation formula, and sets the matching degree higher than the preset matching degree threshold for the available meeting rooms Recommend to users, which simplifies the meeting room reservation process and improves the intelligence of meeting room reservation.
  • the second embodiment of the present application proposes a method for booking a meeting room based on speech recognition.
  • the matching degree is higher than a preset matching degree threshold.
  • the corresponding available meeting room is displayed on the meeting room recommendation interface for users to choose after the steps include:
  • Step S105 Obtain the available meeting room and the scheduled time selected by the user, and mark the available meeting room selected by the user as a target scheduled meeting room;
  • the available meeting room selected by the user on the meeting room recommendation interface is received, and the scheduled time is determined, because the scheduled time may be inconsistent with the start time and/or the meeting duration in the parameters of the meeting room to be booked , It needs to be reconfirmed.
  • the predetermined time may be independently set by the user, or may be obtained after calculation according to the start time and the duration of the meeting.
  • Step S106 Mark the use state of the target scheduled conference room within the predetermined time as in use.
  • the usage status of the target scheduled conference room within the predetermined time before being reserved is idle, so after the user confirms the reservation, the usage status within the predetermined time is modified to the usage status.
  • step S106 after the step of marking the usage status of the target reserved conference room within the predetermined time as in use, the method further includes:
  • a user when a user reserves a conference room, he may need to unsubscribe. Therefore, receive the conference room unsubscription instruction sent by the user through voice or touch commands, and obtain the unsubscribed conference room from the conference room unsubscribe instruction; then use the unsubscribed conference room during the unsubscribe time period The status is changed to idle.
  • the conference room may need to be maintained, there are still conference rooms in the maintenance state.
  • the use state of the conference room in the maintenance state is changed to idle, so that For users to book.
  • the available meeting room and the scheduled time selected by the user are obtained, and the available meeting room selected by the user is marked as a target scheduled meeting room; the target scheduled meeting room is set within the scheduled time
  • the use status is marked as in use, which simplifies the conference room reservation process and improves the intelligence of the conference room reservation.
  • FIG. 4 is a schematic diagram of the functional modules of the first embodiment of the apparatus for booking a meeting room based on voice recognition in this application.
  • the voice recognition-based meeting room reservation device provided by the present application is stored in the memory 1005 of the voice recognition-based meeting room reservation device shown in FIG. 1 to implement all functions of the computer-readable instructions for the voice recognition-based meeting room reservation: Used to receive the conference room reservation voice sent by the user; used to recognize the conference room reservation voice and obtain the conference room parameters to be reserved; used to calculate the conference room parameters to be reserved and the available conference room parameters based on the preset matching degree calculation formula The matching degree between the matching degree; used to display the corresponding available meeting room with the matching degree higher than the preset matching degree threshold on the meeting room recommendation interface for the user to select.
  • the apparatus for booking a conference room based on voice recognition in this embodiment includes:
  • the receiving module 10 is used to receive the conference room reservation voice sent by the user;
  • the voice recognition module 20 is configured to recognize the reserved voice of the conference room and obtain the parameters of the conference room to be reserved;
  • the calculation module 30 is configured to calculate the matching degree between the parameters of the conference room to be reserved and the parameters of the available conference room based on a preset matching degree calculation formula
  • the recommendation module 40 is configured to display the corresponding available meeting rooms with the matching degree higher than the preset matching degree threshold on the meeting room recommendation interface for the user to select;
  • the voice recognition module is also used for:
  • Preprocessing the predetermined voice of the conference room cut off the mute at the beginning and the end of the predetermined voice of the conference room, and divide the predetermined voice of the conference room into frames to obtain multiple frames;
  • a plurality of the keywords are respectively filled in a preset meeting room parameter template to be booked to obtain the meeting room parameter to be booked.
  • the receiving module is also used for:
  • the receiving module is also used for:
  • voice recognition module is also used for:
  • voiceprint information is consistent with the pre-stored voiceprint information, then perform the steps: recognize the scheduled voice of the conference room to obtain the parameters of the conference room to be reserved; if the voiceprint information is inconsistent with the pre-stored voiceprint information, then Ignore the scheduled voice of the conference room.
  • calculation module is also used for:
  • Pre-save the conference room parameters and usage status of all conference rooms mark the conference rooms whose usage status is free within a certain period of time as available conference rooms, and mark the conference room parameters of the available conference rooms as available Meeting room parameters, wherein the specific time period is determined by a preset time floating range and the start time and duration in the predetermined parameters of the meeting room.
  • calculation module is also used for:
  • M a (F a, S a, L a, N a);
  • C represents the matching degree
  • ⁇ F , ⁇ S , ⁇ L , ⁇ N respectively represent the floor weighting factor, the starting time weighting factor, the meeting duration weighting factor, and the number of people weighting factors, ⁇ F , ⁇ S , ⁇ L , ⁇
  • the value of N is preset according to the floor difference, the start time difference, the meeting duration difference, and the number of people difference in the parameters of the meeting room to be reserved and the parameters of the available meeting room respectively;
  • the present application also proposes a computer storage medium that stores a computer readable instruction for booking a meeting room based on voice recognition when the computer readable instruction for booking a meeting room based on voice recognition is executed by a processor The steps of the method for booking a meeting room based on voice recognition as described above are implemented.
  • the computer-readable storage medium may be a non-volatile readable storage medium.
  • the present application proposes a method, device, equipment and storage medium for meeting room reservation based on voice recognition.
  • the method includes: receiving a meeting room reservation voice sent by a user; recognizing the meeting room reservation voice to obtain Meeting room parameters to be booked; calculating the matching degree between the meeting room parameters to be booked and the available meeting room parameters based on the preset matching degree calculation formula; displaying the corresponding available meeting rooms whose matching degree is higher than the preset matching degree threshold Recommend the interface in the meeting room for users to choose.
  • this application obtains the meeting room parameters to be booked based on speech recognition, and then obtains the matching degree between the meeting room parameters to be booked and the available meeting room parameters according to the matching degree calculation formula, and the matching degree is higher than the preset
  • the available meeting rooms that match the threshold are recommended to users, thereby simplifying the meeting room reservation process and improving the intelligence of meeting room reservations.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Telephonic Communication Services (AREA)

Abstract

一种基于语音识别的会议室预定方法、装置、设备及存储介质，该方法包括：接收用户发送的会议室预定语音(S101)；识别所述会议室预定语音，获得待预定会议室参数(S102)；基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度(S103)；将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择(S104)。该方法在人工智能的基础上，基于语音识别获得待预定会议室参数，再根据匹配度计算公式获得待预定会议室参数与可用会议室参数之间的匹配度，并将匹配度高于预设匹配度阈值的可用会议室推荐给用户，由此简化了会议室预定流程，提高了会议室预定的智能性。

Description

基于语音识别的会议室预定方法、装置、设备及存储介质
本申请要求于2019年5月21日提交中国专利局、申请号为201910433354.8、发明名称为“基于语音识别的会议室预定方法、装置、设备及计算机可读存储介质”的中国专利申请的优先权，其全部内容通过引用结合在申请中
技术领域
本申请涉及人工智能技术领域，尤其涉及一种基于语音识别的会议室预定方法、装置、设备及存储介质。
背景技术
当前的会议室预定方法需要用户根据开会时间、与会人数等信息自行挑选可用的会议室，因此，会议室预定流程复杂，用户难以订到符合需要的会议室，会议室使用效率不高。
发明内容
本申请提供一种基于语音识别的会议室预定方法、装置、设备及存储介质，旨在简化会议室预定流程，提高会议室预定的智能性。
为实现上述目的，本申请提供一种基于语音识别的会议室预定方法，所述方法包括：
接收用户发送的会议室预定语音；
识别所述会议室预定语音，获得待预定会议室参数；
基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；
将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择；
所述识别所述会议室预定语音，获得待预定会议室参数的步骤包括：
对所述会议室预定语音进行预处理，切除所述会议室预定语音中首尾端的静音，并将所述会议室预定语音进行分帧，获得多个帧；
提取多个所述帧中的特征向量，并获得与所述特征向量对应的文字；
获取所述文字中的多个关键字；
将多个所述关键字分别填入预设的待预定会议室参数模板中，获得所述待预定会议室参数。
为实现上述目的，本申请还提供一种基于语音识别的会议室预定装置，所述基于语音识别的会议室预定装置包括：
接收模块，用于接收用户发送的会议室预定语音；
语音识别模块，用于识别所述会议室预定语音，获得待预定会议室参数；
计算模块，用于基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；
推荐模块，用于将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择；
所述语音识别模块还用于：
对所述会议室预定语音进行预处理，切除所述会议室预定语音中首尾端的静音，并将所述会议室预定语音进行分帧，获得多个帧；
提取多个所述帧中的特征向量，并获得与所述特征向量对应的文字；
获取所述文字中的多个关键字；
将多个所述关键字分别填入预设的待预定会议室参数模板中，获得所述待预定会议室参数。
为实现上述目的本申请还提供一种基于语音识别的会议室预定设备，所述基于语音识别的会议室预定设备包括处理器，存储器以及存储在所述存储器中的基于语音识别的会议室预定计算机可读指令，所述基于语音识别的会议室预定计算机可读指令被所述处理器运行时，实现如上所述的基于语音识别的会议室预定方法的步骤。
为实现上述目的本申请还提供一种计算机存储介质，所述计算机存储介质上存储有基于语音识别的会议室预定计算机可读指令，所述基于语音识别的会议室预定计算机可读指令被处理器运行时实现如上所述基于语音识别的会议室预定方法的步骤。
相比现有技术，本申请提出的一种基于语音识别的会议室预定方法、装置、设备及存储介质，该方法包括：接收用户发送的会议室预定语音； 识别所述会议室预定语音，获得待预定会议室参数；基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择。本申请在人工智能的基础上，基于语音识别获得待预定会议室参数，再根据匹配度计算公式获得待预定会议室参数与可用会议室参数之间的匹配度，并将匹配度高于预设匹配度阈值的可用会议室推荐给用户，由此简化了会议室预定流程，提高了会议室预定的智能性。
附图说明
图1是本申请各实施例涉及的基于语音识别的会议室预定设备的硬件结构示意图；
图2是本申请基于语音识别的会议室预定方法第一实施例的流程示意图；
图3是本申请基于语音识别的会议室预定方法第二实施例的流程示意图；
图4是本申请基于语音识别的会议室预定装置第一实施例的功能模块示意图。
本申请目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
具体实施方式
应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
本申请实施例主要涉及的基于语音识别的会议室预定设备是指能够实现网络连接的网络连接设备，所述基于语音识别的会议室预定设备可以是服务器、云平台等。
参照图1，图1是本申请各实施例涉及的基于语音识别的会议室预定设备的硬件结构示意图。本申请实施例中，基于语音识别的会议室预定设备可以包括处理器1001(例如中央处理器Central Processing Unit、CPU)，通信总线1002，输入端口1003，输出端口1004，存储器1005。其中，通信 总线1002用于实现这些组件之间的连接通信；输入端口1003用于数据输入；输出端口1004用于数据输出，存储器1005可以是高速RAM存储器，也可以是稳定的存储器(non-volatile memory)，例如磁盘存储器，存储器1005可选的还可以是独立于前述处理器1001的存储装置。本领域技术人员可以理解，图1中示出的硬件结构并不构成对本申请的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
继续参照图1，图1中作为一种可读存储介质的存储器1005可以包括操作系统、网络通信模块、应用程序模块以及基于语音识别的会议室预定计算机可读指令。在图1中，网络通信模块主要用于连接服务器，与服务器进行数据通信；而处理器1001可以调用存储器1005中存储的基于语音识别的会议室预定计算机可读指令，并执行本申请实施例提供的基于语音识别的会议室预定方法。
本申请实施例提供了一种基于语音识别的会议室预定方法。
参照图2，图2是本申请基于语音识别的会议室预定方法第一实施例的流程示意图。
本实施例中，所述基于语音识别的会议室预定方法应用于基于语音识别的会议室预定设备，所述方法包括：
步骤S101，接收用户发送的会议室预定语音；
当前很多会议室的预定都需要人工记录预定情况，即使有些公司开发了线上预定会议室的产品，但仍然用户需要在会议室预定过程中，在预设的会议室配置界面输入所需要会议室的详细信息，例如会议室编号、占用时间等。
本实施例中，利用基于语音识别的会议室预定设备的录音功能，接收用户的会议室预定语音，所述会议室预定语音包括待预定会议室参数，所述待预定会议室参数包括楼层、起始时间、会议持续时长、与会人数中的一个或多个。
进一步地，所述接收用户发送的会议室预定语音的步骤之前还包括:
通过基于语音识别的会议室预定设备的显示屏显示语音输入提示，以供用户参考所述语音输入提示进行语音输入。
当检测到用户需要输入会议室预定语音时，通过所述基于语音识别的会议室预定设备的显示屏显示语音输入提示，或者以语音的方式播放预先 录制的语音输入提示文件。参例如所述语音输入提示可以是：我需要预定今天下午X点到Y点XX楼X人的会议室。
步骤S102，识别所述会议室预定语音，获得待预定会议室参数；
本实施例中，所述识别所述会议室预定语音，获得待预定会议室参数的步骤包括：
步骤S102-1，对所述会议室预定语音进行预处理，切除所述会议室预定语音中首尾端的静音，并将所述会议室预定语音进行分帧，获得多个帧；
首先获取所述会议室预定语音中的静音：将所述会议室预定语音分成固定长度的语音，并形成一个语音序列，提取所述语音序列中每一段语音的短时能量、短时过零率和短时信息熵三个音频特征参数；根据音频特征参数计算每一帧数据的短时能频值，并形成一个短时能频值序列；从第一段语音开始，分析短时能频值序列，并找出每一段语音的起点和终点；将所述起点和所述终点之外的语音标记为静音期，所述静音期内的语音即为静音。所述静音包括所述会议室预定语音中首尾端的静音，由此通过消除所述静音以达到在不降低业务质量的情况下节省资源并降低干扰的作用。进一步地，将切除了静音的所述会议室预定语音切开成一小段一小段，例如每25ms为一小段，每一小段称为一帧，由此分帧后可获得多个帧，一般地，可以使用移动窗函数来实现，相邻的帧之间一般有相互交叠的部分。若每帧的长度为25ms，可以在相邻的帧之间设置15ms的交叠，如此，以帧长25ms，帧移10ms进行分帧。
步骤S102-2，提取多个所述帧中的特征向量，并获得与所述特征向量对应的文字；
分帧后，所述会议室预定语音变成了多个帧，但是波型在时域上不具有描述能力，因此必须进一步进行波型变换。具体地，对所述多个帧进行加窗，获得多个短时分析窗，得到对应的频谱；并用滤波器对所述频谱进行滤波获得滤波频谱；对所述滤波频谱通过取对数、逆变换的方式进行倒谱分析，获得获得所述会议室预定语音的特征向量。由此，根据所述会议室预定语音中的声学特征将所述会议室预定语音中的每一帧变成一个多维的特征向量，所述特征向量包括该帧会议室预定语音的内容信息。
所述获得与所述特征向量对应的文字的步骤包括：
具体地，预先提取大量训练语音中的特征向量，获得训练特征向量， 利用所述训练特征向量进行声学模型训练。所述声学模型可以输出与所述特征向量对应的音素信息。从字典中查询与所述音素信息对应的字和/或词。将所述字和/或词输入预先训练的所述语言模型，得到与所述字和/或词相互关联的概率，再基于所述概率，输出语序合理的文字，所述文字与所述特征向量对应。其中所述语言模型是通过对大量文本信息进行训练后得到的，利用所述语言模型可以获得单个字或词相互关联的概率。
步骤S102-3，获取所述文字中的多个关键字；
具体地，获取与会议室预定相关的关键字，并去除多余的文字。预先设置关键字库，所述关键字库包括数字、时间、楼层等与会议室预定有关的关键字。将所述文字中与所述关键字库中相同的字和\或词标记为关键字。可以理解地，所述关键字有多个。
步骤S102-4，将多个所述关键字分别填入预设的待预定会议室参数模板中，获得所述待预定会议室参数。
本实施例中，预先设置待预定会议室参数模板，所述待预定会议室参数模板中包括多个所述关键字的填入位置，以便填入所述关键字，获得所述待预定会议室参数。例如所述待预定会议室参数模板可以设置为：楼层是(X F)，起始时间是(X S)，持续时长是(X L)，与会人数是(X N)；其中括号内的X F、X S、X L以及X N为相应内容的数值。
进一步地，所述识别所述会议室预定语音，获得待预定会议室参数的步骤之前还包括：
步骤S102a：提取所述会议室预定语音中的声纹信息，将所述声纹信息与预存声纹信息进行对比；
本实施例中，预先采集一个或多个预定声纹信息，为所述预定声纹信息配置会议室预定权限，并记录所述会议室预定声纹信息对应用户的详细信息，例如姓名、联系电话、邮箱、职位等。当接收到用户发送的会议室预定语音后，提取所述会议室预定语音中的声纹信息，并将所述声纹信息与预存声纹信息进行对比；
若所述声纹信息与所述预存声纹信息一致，则判定所述声纹信息对应的用户具有会议室预定权重，则执行步骤：识别所述会议室预定语音，获得待预定会议室参数；若所述声纹信息与所述预存声纹信息不一致，则判定所述声纹信息对应的用户不具有会议室预定权重，则忽略所述会议室预 定语音。由此，可以防止不具有权限的用户进行会议室预定，便于会议室预定的管理。
步骤S103，基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；
本实施例中，所述基于预设匹配度算法计算所述待预定会议室参数与可用会议室参数之间的匹配度的步骤之前还包括：预先保存所有会议室的会议室参数和使用状态，将所述使用状态在特定时间段内为空闲的会议室标记为可用会议室，并将所述可用会议室的会议室参数标记为所述可用会议室参数，其中，所述特定时间段由预先设定的时间浮动范围以及所述会议室预定参数中的起始时间和持续时间决定。由此，预先获得本公司、本建筑内的所有会议室的所述会议室参数，并将所有会议室的所述会议室参数保存至数据库，以便调用。可以理解地，所述使用状态包括使用中、空闲以及维护状态。对于会议室而言，所述使用状态又与时间相关，故将在特定时间段内为空闲的会议室标记为可用会议室。其中，所述特定时间段由预先设定的时间浮动范围以及所述会议室预定参数中的起始时间和持续时间决定。所述时间浮动范围可以具体设置，例如所述起始时间的前后的一定时长，例如60min、120min等；或者所述起始时间点的向前或向后移动预设天数，例如起始时间点向前一天或往后若干天。
本实施例中，所述基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度的步骤包括：
步骤a：将所述待预定会议室M a表示为：M a＝(F a,S a,L a,N a)；
步骤b：将所述可用会议室M r表示为：M r＝(F r,S r,L r,N r)；其中，F a,S a,L a,N a为所述待预定会议室参数，F r,S r,L r,N r为所述可用会议室参数，F表示楼层，S表示起始时间，L表示会议持续时长，N表示与会人数；
可以理解地，所述待预定会议室参数和所述可用会议室参数可以包括比F、S、L、N更多或更少的参数。
步骤c：将所述匹配度计算公式表示为：
Figure PCTCN2019122064-appb-000001
其中，C表示匹配度，η F,η S,η L,η N分别表示楼层权重因子、起始时间权重因子、会议持续时长权重因子、人数权重因子，η F,η S,η L,η N的值分别根据所述待预定会议室参数与所述可用会议室参数中的楼层差、起始时间差、会 议持续时长差以及人数差值预先设置。可以理解地，当某个可用会议室参数与所述待预定会议室参数完全一致时，所述匹配度C为100；当某个可用会议室参数与所述待预定会议室参数的差别越大，所述匹配度C越小。
进一步地，所述η F、η S、η L、η N的设置方法如下：
Figure PCTCN2019122064-appb-000002
其中，x 1,x 2,x 3,x 4,x 5表示所述楼层差，本实施例中x 1＜x 2＜x 3＜x 4＜x 5，均为正整数，根据所述楼层差确定所述楼层权重因子η F的取值：当所述楼层差在[0,x 1]之间，所述η F的取值为η F1；当所述楼层差在[x 2,x 3]之间，所述η F的取值为η F2；当所述楼层差在[x 4,x 5]之间，所述η F的取值为η F3；其中η F1,η F2,η F3的值可以根据经验设置，例如分别设置为0.8、1.0以及1.2。
Figure PCTCN2019122064-appb-000003
其中，y 1,y 2,y 3,y 4,y 5表示所述起始时间差，并且y 1＜y 2＜y 3＜y 4＜y 5，根据所述起始时间差确定所述起始时间权重因子η S的取值：当所述楼层差在[0,y 1]之间，所述η S的取值为η S1；当所述楼层差在[y 2,y 3]之间，所述η S的取值为η S2；当所述楼层差在[y 4,y 5]之间，所述η S的取值为η S3；其中，η S1,η S2,η S3的值可以根据经验设置，例如分别设置为0.8、1.0以及1.2。
Figure PCTCN2019122064-appb-000004
其中，z 1,z 2,z 3,z 4,z 5表示所述会议持续时长差，并且z 1＜z 2＜z 3＜z 4＜z 5，根据所述会议持续时长差确定所述会议持续时长因子η L的取值：当所述会议持续时长差在[0,z 1]之间，所述η L的取值为η L1；当所述会议持续时长差在[z 2,z 3]之间，所述η L的取值为η L2；当所述会议持续时长差在[z 4,z 5]之间，所述η L的取值为η L3；其中，η L1,η L2,η L3的值可以根据需要具体设置，例如分别设置为0.8、1.0以及1.2。
Figure PCTCN2019122064-appb-000005
其中，m 1,m 2,m 3,m 4,m 5表示所述人数差，m 1,m 2,m 3,m 4,m 5均为正整数，并且m 1＜m 2＜m 3＜m 4＜m 5，根据所述人数差确定所述人数权重因子η N的取值：当所述人数差在[0,m 1]之间，所述η N的取值为η N1，当所述人数差在[m 2,m 3]之间，所述η N的取值为η N2，当所述人数差在[m 4,m 5]之间，所述η N的取值为η N3；其中，η N1，η N2，η N3的值可以根据需要具体设置，例如分别设置为0.8、1.0以及1.2。
例如，将A公司的η F、η S、η L、η N分别设置如下：
Figure PCTCN2019122064-appb-000006
Figure PCTCN2019122064-appb-000007
Figure PCTCN2019122064-appb-000008
Figure PCTCN2019122064-appb-000009
步骤d：将所述权重因子、所述可用会议室参数、所述待预定会议室参数代入所述匹配度计算公式计算所述匹配度。
具体地，再根据所述待预定会议室参数和所述可用会议室参数获得对应的η F、η S、η L、η N；再根据匹配度计算公式即可获得所述可用会议室参数、所述待预定会议室参数之间的匹配度，可以理解地，所述匹配度所述可用会议室参数、所述待预定会议室参数与各个可用会议室与待预定会议室对应。由于所述可用会议室有一个或多个，故所述匹配度也对应有一个或多个。
步骤S104，将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择。
本实施例中，预先设置匹配度阈值，例如80％。若某个可用会议室参数与所述待预定会议室参数完全一致时，所述匹配度C为100；若某个可用会议室参数与所述待预定会议室参数的差别越小，所述匹配度C越大；若某个可用会议室参数与所述待预定会议室参数的差别越大，所述匹配度C越小。因此，本实施例将所述匹配度高于预设匹配度阈值的所述对应可用会议室显示在会议室推荐界面，以供用户选择。
进一步地，还可以设置排队功能，接收用户发送的排队指令，当所述可用会议室中有匹配度更高的可用会议室时，则将所述匹配度更高的会议室推荐给用户。
本实施例通过上述方案，接收用户发送的会议室预定语音；识别所述会议室预定语音，获得待预定会议室参数；基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择。本申请基于语音识别获得待预定会议室参数，再根据匹配度计算公式获得待预定会议室参数与可用会议室参数之间的匹配度，并将匹配度高于预设匹配度阈值的可用会议室推荐给用户，由此简化了会议室预定流程， 提高了会议室预定的智能性。
如图3所示，本申请第二实施例提出一种基于语音识别的会议室预定方法，基于上述图2所示的第一实施例，所述将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择的步骤之后还包括：
步骤S105，获取用户选择的所述可用会议室以及预定时间，将用户选择的所述可用会议室标记为目标预定会议室；
本实施例中，接收用户在所述会议室推荐界面选择的可用会议室，确定所述预定时间，由于所述预定时间可能与待预定会议室参数中的起始时间和\或会议持续时长不一致，故需要重新进行确认。所述预定时间可以由用户自主设置，也可以根据所述起始时间和所述会议持续时长计算后获得。
待接收到用户发送的确认指令后，将用户选择的所述可用会议室标记为目标预定会议室。
步骤S106，将所述目标预定会议室在所述预定时间内的使用状态标记为使用中。
具体地，所述目标预定会议室在被预定之前在所述预定时间内的使用状态为空闲，故在用户确定预定之后，将所述预定时间内的使用状态修改为使用状态。
进一步地，所述步骤S106：将所述目标预定会议室在所述预定时间内的使用状态标记为使用中的步骤之后还包括：
接收用户发送的会议室退订指令，从所述会议室退订指令获取待退订会议室；
将所述待退订会议室在退订时间段内的使用状态修改为空闲。
具体地，当用户预定会议室后，可能需要退订。因此接收用户通过语音或触控指令发送的会议室退订指令，并从所述会议室退订指令获取待退订会议室；然后将所述待退订会议室在退订时间段内的使用状态修改为空闲。
可以理解地，由于会议室可能需要维护故还有会议室处于维护状态，当处于维护状态的所述会议室在维护完成后，则将所述维护状态的会议室的使用状态修改为空闲，以供用户预定。
本实施例通过上述方案，获取用户选择的所述可用会议室以及预定时 间，将用户选择的所述可用会议室标记为目标预定会议室；将所述目标预定会议室在所述预定时间内的使用状态标记为使用中，由此简化了会议室预定流程，提高了会议室预定的智能性。
此外，本实施例还提供一种基于语音识别的会议室预定装置。参照图4，图4为本申请基于语音识别的会议室预定装置第一实施例的功能模块示意图。
本申请提供的基于语音识别的会议室预定装置，存储于图1所示的基于语音识别的会议室预定设备的存储器1005中，以实现基于语音识别的会议室预定计算机可读指令的所有功能：用于接收用户发送的会议室预定语音；用于识别所述会议室预定语音，获得待预定会议室参数；用于基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；用于将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择。
具体地，本实施例中所述基于语音识别的会议室预定装置包括：
接收模块10，用于接收用户发送的会议室预定语音；
语音识别模块20，用于识别所述会议室预定语音，获得待预定会议室参数；
计算模块30，用于基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；
推荐模块40，用于将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择；
所述语音识别模块还用于：
对所述会议室预定语音进行预处理，切除所述会议室预定语音中首尾端的静音，并将所述会议室预定语音进行分帧，获得多个帧；
提取多个所述帧中的特征向量，并获得与所述特征向量对应的文字；
获取所述文字中的多个关键字；
将多个所述关键字分别填入预设的待预定会议室参数模板中，获得所述待预定会议室参数。
进一步地，所述接收模块还用于：
获取用户选择的所述可用会议室以及预定时间，将用户选择的所述可 用会议室标记为目标预定会议室；
将所述目标预定会议室在所述预定时间内的使用状态标记为使用中。
进一步地，所述接收模块还用于：
接收用户发送的会议室退订指令，从所述会议室退订指令获取待退订会议室；
将所述待退订会议室在退订时间段内的使用状态修改为空闲。
进一步地，所述语音识别模块还用于：
提取所述会议室预定语音中的声纹信息，将所述声纹信息与预存声纹信息进行对比；
若所述声纹信息与所述预存声纹信息一致，则执行步骤：识别所述会议室预定语音，获得待预定会议室参数；若所述声纹信息与所述预存声纹信息不一致，则忽略所述会议室预定语音。
进一步地，所述计算模块还用于：
预先保存所有会议室的会议室参数和使用状态，将所述使用状态在特定时间段内为空闲的会议室标记为可用会议室，并将所述可用会议室的会议室参数标记为所述可用会议室参数，其中，所述特定时间段由预先设定的时间浮动范围以及所述会议室预定参数中的起始时间和持续时间决定。
进一步地，所述计算模块还用于：
将所述待预定会议室M a表示为：M a＝(F a,S a,L a,N a)；
将所述可用会议室M r表示为：M r＝(F r,S r,L r,N r)；其中，F a,S a,L a,N a为所述待预定会议室参数，F r,S r,L r,N r为所述可用会议室参数，F表示楼层，S表示起始时间，L表示会议持续时长，N表示与会人数；
将所述匹配度计算公式表示为：
Figure PCTCN2019122064-appb-000010
其中，C表示匹配度，η F,η S,η L,η N分别表示楼层权重因子、起始时间权重因子、会议持续时长权重因子、人数权重因子，η F,η S,η L,η N的值分别根据所述待预定会议室参数与所述可用会议室参数中的楼层差、起始时间差、会议持续时长差以及人数差值预先设置；
将所述权重因子、所述可用会议室参数、所述待预定会议室参数代入所述匹配度计算公式计算所述匹配度。
此外，本申请还提出一种计算机存储介质，所述计算机存储介质上存 储有基于语音识别的会议室预定计算机可读指令，所述基于语音识别的会议室预定计算机可读指令被处理器运行时实现如上所述的基于语音识别的会议室预定方法的步骤。所述计算机可读存储介质可以为非易失性可读存储介质。
相比现有技术，本申请提出的一种基于语音识别的会议室预定方法、装置、设备及存储介质，该方法包括：接收用户发送的会议室预定语音；识别所述会议室预定语音，获得待预定会议室参数；基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择。本申请在人工智能的基础上，基于语音识别获得待预定会议室参数，再根据匹配度计算公式获得待预定会议室参数与可用会议室参数之间的匹配度，并将匹配度高于预设匹配度阈值的可用会议室推荐给用户，由此简化了会议室预定流程，提高了会议室预定的智能性。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者系统不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者系统所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者系统中还存在另外的相同要素。
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在如上所述的一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备执行本申请各个实施例所述的方法。
以上所述仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或流程变换，或直接或间接运用在其它相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (20)

  1. 一种基于语音识别的会议室预定方法，其中，所述方法包括：
    接收用户发送的会议室预定语音；
    识别所述会议室预定语音，获得待预定会议室参数；
    基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；
    将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择；
    所述识别所述会议室预定语音，获得待预定会议室参数的步骤包括：
    对所述会议室预定语音进行预处理，切除所述会议室预定语音中首尾端的静音，并将所述会议室预定语音进行分帧，获得多个帧；
    提取多个所述帧中的特征向量，并获得与所述特征向量对应的文字；
    获取所述文字中的多个关键字；
    将多个所述关键字分别填入预设的待预定会议室参数模板中，获得所述待预定会议室参数。
  2. 根据权利要求1所述的方法，其中，所述将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择的步骤之后还包括：
    获取用户选择的所述可用会议室以及预定时间，将用户选择的所述可用会议室标记为目标预定会议室；
    将所述目标预定会议室在所述预定时间内的使用状态标记为使用中。
  3. 根据权利要求2所述的方法，其中，所述将所述目标预定会议室在所述预定时间内的使用状态标记为使用中的步骤之后还包括：
    接收用户发送的会议室退订指令，从所述会议室退订指令获取待退订会议室；
    将所述待退订会议室在退订时间段内的使用状态修改为空闲。
  4. 根据权利要求1所述的方法，其中，所述识别所述会议室预定语音，获得待预定会议室参数的步骤之前还包括：
    提取所述会议室预定语音中的声纹信息，将所述声纹信息与预存声纹信息进行对比；
    若所述声纹信息与所述预存声纹信息一致，则执行步骤：识别所述会议室预定语音，获得待预定会议室参数；
    若所述声纹信息与所述预存声纹信息不一致，则忽略所述会议室预定语音。
  5. 根据权利要求1所述的方法，其中，所述基于预设匹配度算法计算所述待预定会议室参数与可用会议室参数之间的匹配度的步骤之前还包括：
    预先保存所有会议室的会议室参数和使用状态，将所述使用状态在特定时间段内为空闲的会议室标记为可用会议室，并将所述可用会议室的会议室参数标记为所述可用会议室参数，其中，所述特定时间段由预先设定的时间浮动范围以及所述会议室预定参数中的起始时间和持续时间决定。
  6. 根据权利要求1所述的方法，其中，所述基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度的步骤包括：
    将所述待预定会议室M a表示为：M a＝(F a,S a,L a,N a)；
    将所述可用会议室M r表示为：M r＝(F r,S r,L r,N r)；其中，F a,S a,L a,N a为所述待预定会议室参数，F r,S r,L r,N r为所述可用会议室参数，F表示楼层，S表示起始时间，L表示会议持续时长，N表示与会人数；
    将所述匹配度计算公式表示为：
    Figure PCTCN2019122064-appb-100001
    其中，C表示匹配度，η F,η S,η L,η N分别表示楼层权重因子、起始时间权重因子、会议持续时长权重因子、人数权重因子，η F,η S,η L,η N的值分别根据所述待预定会议室参数与所述可用会议室参数中的楼层差、起始时间差、会议持续时长差以及人数差值预先设置；
    将所述权重因子、所述可用会议室参数、所述待预定会议室参数代入所述匹配度计算公式计算所述匹配度。
  7. 一种基于语音识别的会议室预定装置，其中，所述基于语音识别的会议室预定装置包括：
    接收模块，用于接收用户发送的会议室预定语音；
    语音识别模块，用于识别所述会议室预定语音，获得待预定会议室参数；
    计算模块，用于基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；
    推荐模块，用于将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择
    所述语音识别模块还用于：
    对所述会议室预定语音进行预处理，切除所述会议室预定语音中首尾端的静音，并将所述会议室预定语音进行分帧，获得多个帧；
    提取多个所述帧中的特征向量，并获得与所述特征向量对应的文字；
    获取所述文字中的多个关键字；
    将多个所述关键字分别填入预设的待预定会议室参数模板中，获得所述待预定会议室参数。
  8. 根据权利要求7所述的装置，其中，所述推荐模块还用于：
    获取用户选择的所述可用会议室以及预定时间，将用户选择的所述可用会议室标记为目标预定会议室；
    将所述目标预定会议室在所述预定时间内的使用状态标记为使用中。
  9. 根据权利要求8所述的装置，其中，所述推荐模块还用于：
    接收用户发送的会议室退订指令，从所述会议室退订指令获取待退订会议室；
    将所述待退订会议室在退订时间段内的使用状态修改为空闲。
  10. 根据权利要求7所述的装置，其中，所述语音识别模块还用于：
    提取所述会议室预定语音中的声纹信息，将所述声纹信息与预存声纹信息进行对比；
    若所述声纹信息与所述预存声纹信息一致，则执行步骤：识别所述会议室预定语音，获得待预定会议室参数；
    若所述声纹信息与所述预存声纹信息不一致，则忽略所述会议室预定语音。
  11. 根据权利要求7所述的装置，其中，所述计算模块还包括：
    预先保存所有会议室的会议室参数和使用状态，将所述使用状态在特定时间段内为空闲的会议室标记为可用会议室，并将所述可用会议室的会议室参数标记为所述可用会议室参数，其中，所述特定时间段由预先设定的时间浮动范围以及所述会议室预定参数中的起始时间和持续时间决定。
  12. 根据权利要求7所述的装置，其中，所述计算模块还用于：
    将所述待预定会议室M a表示为：M a＝(F a,S a,L a,N a)；
    将所述可用会议室M r表示为：M r＝(F r,S r,L r,N r)；其中，F a,S a,L a,N a为所述待预定会议室参数，F r,S r,L r,N r为所述可用会议室参数，F表示楼层，S表示起始时间，L表示会议持续时长，N表示与会人数；
    将所述匹配度计算公式表示为：
    Figure PCTCN2019122064-appb-100002
    其中，C表示匹配度，η F,η S,η L,η N分别表示楼层权重因子、起始时间权重因子、会议持续时长权重因子、人数权重因子，η F,η S,η L,η N的值分别根据所述待预定会议室参数与所述可用会议室参数中的楼层差、起始时间差、会议持续时长差以及人数差值预先设置；
    将所述权重因子、所述可用会议室参数、所述待预定会议室参数代入所述匹配度计算公式计算所述匹配度。
  13. 一种基于语音识别的会议室预定设备，其中，所述基于语音识别的会议室预定设备包括处理器，存储器以及存储在所述存储器中的基于语音识别的会议室预定计算机可读指令，所述基于语音识别的会议室预定计算机可读指令被所述处理器运行时，实现如下步骤：
    接收用户发送的会议室预定语音；
    识别所述会议室预定语音，获得待预定会议室参数；
    基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；
    将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择；
    所述识别所述会议室预定语音，获得待预定会议室参数的步骤包括：
    对所述会议室预定语音进行预处理，切除所述会议室预定语音中首尾端的静音，并将所述会议室预定语音进行分帧，获得多个帧；
    提取多个所述帧中的特征向量，并获得与所述特征向量对应的文字；
    获取所述文字中的多个关键字；
    将多个所述关键字分别填入预设的待预定会议室参数模板中，获得所述待预定会议室参数。
  14. 根据权利要求13所述的设备，其中，所述将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择的步骤之后还包括：
    获取用户选择的所述可用会议室以及预定时间，将用户选择的所述可用会议室标记为目标预定会议室；
    将所述目标预定会议室在所述预定时间内的使用状态标记为使用中。
  15. 根据权利要求14所述的设备，其中，所述将所述目标预定会议室在所述预定时间内的使用状态标记为使用中的步骤之后还包括：
    接收用户发送的会议室退订指令，从所述会议室退订指令获取待退订会议室；
    将所述待退订会议室在退订时间段内的使用状态修改为空闲。
  16. 根据权利要求13所述的设备，其中，所述识别所述会议室预定语音，获得待预定会议室参数的步骤之前还包括：
    提取所述会议室预定语音中的声纹信息，将所述声纹信息与预存声纹信息进行对比；
    若所述声纹信息与所述预存声纹信息一致，则执行步骤：识别所述会议室预定语音，获得待预定会议室参数；
    若所述声纹信息与所述预存声纹信息不一致，则忽略所述会议室预定语音。
  17. 根据权利要求13所述的设备，其中，所述基于预设匹配度算法计算所述待预定会议室参数与可用会议室参数之间的匹配度的步骤之前还包括：
    预先保存所有会议室的会议室参数和使用状态，将所述使用状态在特定时间段内为空闲的会议室标记为可用会议室，并将所述可用会议室的会议室参数标记为所述可用会议室参数，其中，所述特定时间段由预先设定的时间浮动范围以及所述会议室预定参数中的起始时间和持续时间决定。
  18. 根据权利要求13所述的设备，其中，所述基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度的步骤包括：
    将所述待预定会议室M a表示为：M a＝(F a,S a,L a,N a)；
    将所述可用会议室M r表示为：M r＝(F r,S r,L r,N r)；其中，F a,S a,L a,N a为所述待预定会议室参数，F r,S r,L r,N r为所述可用会议室参数，F表示楼层，S表示起始时间，L表示会议持续时长，N表示与会人数；
    将所述匹配度计算公式表示为：
    Figure PCTCN2019122064-appb-100003
    其中，C表示匹配度，η F,η S,η L,η N分别表示楼层权重因子、起始时间权重因子、会议持续时长权重因子、人数权重因子，η F,η S,η L,η N的值分别根据所述待预定会议室参数与所述可用会议室参数中的楼层差、起始时间差、会议持续时长差以及人数差值预先设置；
    将所述权重因子、所述可用会议室参数、所述待预定会议室参数代入所述匹配度计算公式计算所述匹配度。
  19. 一种计算机存储介质，其中，所述计算机存储介质上存储有基于语音识别的会议室预定计算机可读指令，所述基于语音识别的会议室预定计算机可读指令被处理器运行时实现如下的步骤：
    接收用户发送的会议室预定语音；
    识别所述会议室预定语音，获得待预定会议室参数；
    基于预设匹配度计算公式计算所述待预定会议室参数与可用会议室参数之间的匹配度；
    将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择；
    所述识别所述会议室预定语音，获得待预定会议室参数的步骤包括：
    对所述会议室预定语音进行预处理，切除所述会议室预定语音中首尾端的静音，并将所述会议室预定语音进行分帧，获得多个帧；
    提取多个所述帧中的特征向量，并获得与所述特征向量对应的文字；
    获取所述文字中的多个关键字；
    将多个所述关键字分别填入预设的待预定会议室参数模板中，获得所述待预定会议室参数。
  20. 根据权利要求19所述的计算机存储介质，其中，所述将所述匹配度高于预设匹配度阈值的对应可用会议室显示在会议室推荐界面，以供用户选择的步骤之后还包括：
    获取用户选择的所述可用会议室以及预定时间，将用户选择的所述可用会议室标记为目标预定会议室；
    将所述目标预定会议室在所述预定时间内的使用状态标记为使用中。
PCT/CN2019/122064 2019-05-21 2019-11-29 基于语音识别的会议室预定方法、装置、设备及存储介质 WO2020233075A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910433354.8 2019-05-21
CN201910433354.8A CN110298463A (zh) 2019-05-21 2019-05-21 基于语音识别的会议室预定方法、装置、设备及存储介质

Publications (1)

Publication Number Publication Date
WO2020233075A1 true WO2020233075A1 (zh) 2020-11-26

Family

ID=68027103

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/122064 WO2020233075A1 (zh) 2019-05-21 2019-11-29 基于语音识别的会议室预定方法、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN110298463A (zh)
WO (1) WO2020233075A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113011622A (zh) * 2021-03-16 2021-06-22 北京百度网讯科技有限公司 会议室预定的处理方法及装置、计算设备和介质
CN114202800A (zh) * 2021-11-12 2022-03-18 珠海大横琴科技发展有限公司 一种照明设备的控制方法和装置
CN115496246A (zh) * 2022-10-08 2022-12-20 卓思韦尔(北京)信息技术有限公司 基于群体差异的共享会议室智能搜索及灵动分配方法
CN115630719A (zh) * 2022-10-11 2023-01-20 深圳中正信息科技有限公司 会议室预定方法、系统及计算机可读存储介质

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110298463A (zh) * 2019-05-21 2019-10-01 深圳壹账通智能科技有限公司 基于语音识别的会议室预定方法、装置、设备及存储介质
CN112203038B (zh) * 2020-10-12 2022-09-16 北京字节跳动网络技术有限公司 在线会议的处理方法、装置、电子设备及计算机存储介质
CN114038468B (zh) * 2022-01-07 2022-04-15 深圳市声扬科技有限公司 语音数据比对处理方法、装置、电子设备和存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106934471A (zh) * 2017-02-17 2017-07-07 北京光年无限科技有限公司 应用于智能机器人的会议室管理方法及系统
US9741020B2 (en) * 2010-01-27 2017-08-22 Google Inc. Conference room scheduling based on attendee locations
CN107766958A (zh) * 2017-10-27 2018-03-06 珠海华发新科技投资控股有限公司 会议室预定系统及方法
CN108734320A (zh) * 2018-05-09 2018-11-02 北京邦邦共赢网络科技有限公司 一种办公方法及装置
CN109344988A (zh) * 2018-09-30 2019-02-15 塔米智能科技（北京）有限公司 一种会议管理系统
CN110298463A (zh) * 2019-05-21 2019-10-01 深圳壹账通智能科技有限公司 基于语音识别的会议室预定方法、装置、设备及存储介质

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH11175093A (ja) * 1997-12-08 1999-07-02 Nippon Telegr & Teleph Corp <Ntt> 音声認識確認応答方法
CN109545221B (zh) * 2019-01-23 2024-03-19 努比亚技术有限公司 参数调整方法、移动终端及计算机可读存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9741020B2 (en) * 2010-01-27 2017-08-22 Google Inc. Conference room scheduling based on attendee locations
CN106934471A (zh) * 2017-02-17 2017-07-07 北京光年无限科技有限公司 应用于智能机器人的会议室管理方法及系统
CN107766958A (zh) * 2017-10-27 2018-03-06 珠海华发新科技投资控股有限公司 会议室预定系统及方法
CN108734320A (zh) * 2018-05-09 2018-11-02 北京邦邦共赢网络科技有限公司 一种办公方法及装置
CN109344988A (zh) * 2018-09-30 2019-02-15 塔米智能科技（北京）有限公司 一种会议管理系统
CN110298463A (zh) * 2019-05-21 2019-10-01 深圳壹账通智能科技有限公司 基于语音识别的会议室预定方法、装置、设备及存储介质

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113011622A (zh) * 2021-03-16 2021-06-22 北京百度网讯科技有限公司 会议室预定的处理方法及装置、计算设备和介质
CN114202800A (zh) * 2021-11-12 2022-03-18 珠海大横琴科技发展有限公司 一种照明设备的控制方法和装置
CN115496246A (zh) * 2022-10-08 2022-12-20 卓思韦尔(北京)信息技术有限公司 基于群体差异的共享会议室智能搜索及灵动分配方法
CN115630719A (zh) * 2022-10-11 2023-01-20 深圳中正信息科技有限公司 会议室预定方法、系统及计算机可读存储介质
CN115630719B (zh) * 2022-10-11 2023-09-12 深圳中正信息科技有限公司 会议室预定方法、系统及计算机可读存储介质

Also Published As

Publication number Publication date
CN110298463A (zh) 2019-10-01

Similar Documents

Publication Publication Date Title
WO2020233075A1 (zh) 基于语音识别的会议室预定方法、装置、设备及存储介质
US11417343B2 (en) Automatic speaker identification in calls using multiple speaker-identification parameters
US10522151B2 (en) Conference segmentation based on conversational dynamics
US11682401B2 (en) Matching speakers to meeting audio
US10516782B2 (en) Conference searching and playback of search results
US10567185B2 (en) Post-conference playback system having higher perceived quality than originally heard in the conference
EP3582119A1 (en) Spoken language understanding system and method using recurrent neural networks
US8170866B2 (en) System and method for increasing accuracy of searches based on communication network
DE102017124264B4 (de) Computerimplementiertes Verfahren und Rechensystem zum Bestimmen phonetischer Beziehungen
US20180027351A1 (en) Optimized virtual scene layout for spatial meeting playback
US10860797B2 (en) Generating summaries and insights from meeting recordings
US20180191912A1 (en) Selective conference digest
CN110751943A (zh) 一种语音情绪识别方法、装置以及相关设备
CN107818798A (zh) 客服服务质量评价方法、装置、设备及存储介质
US20180190266A1 (en) Conference word cloud
CN109545197B (zh) 语音指令的识别方法、装置和智能终端
JP2008506156A (ja) マルチスロット対話システムおよび方法
US20170154030A1 (en) Providing electronic text recommendations to a user based on what is discussed during a meeting
US20190384854A1 (en) Generating summaries and insights from meeting recordings
CN112908301A (zh) 一种语音识别方法、装置、存储介质及设备
CN109074809B (zh) 信息处理设备、信息处理方法和计算机可读存储介质
CN111128175A (zh) 口语对话管理方法及系统
Johar Paralinguistic profiling using speech recognition
JP2020077272A (ja) 会話システムおよび会話プログラム
CN112599125A (zh) 一种语音办公处理方法、装置、终端及存储介质

Legal Events

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

Ref document number: 19930079

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

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

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 25.02.2022)

122 Ep: pct application non-entry in european phase

Ref document number: 19930079

Country of ref document: EP

Kind code of ref document: A1