WO2019119927A1 - 语音打车方法及*** - Google Patents

语音打车方法及*** Download PDF

Info

Publication number
WO2019119927A1
WO2019119927A1 PCT/CN2018/109010 CN2018109010W WO2019119927A1 WO 2019119927 A1 WO2019119927 A1 WO 2019119927A1 CN 2018109010 W CN2018109010 W CN 2018109010W WO 2019119927 A1 WO2019119927 A1 WO 2019119927A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
voice
module
taxi
destination
Prior art date
Application number
PCT/CN2018/109010
Other languages
English (en)
French (fr)
Inventor
于治武
袁晓春
赵成旺
Original Assignee
北京京东尚科信息技术有限公司
北京京东世纪贸易有限公司
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 北京京东尚科信息技术有限公司, 北京京东世纪贸易有限公司 filed Critical 北京京东尚科信息技术有限公司
Publication of WO2019119927A1 publication Critical patent/WO2019119927A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • 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
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • 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
    • G10L15/08—Speech classification or search
    • G10L15/10—Speech classification or search using distance or distortion measures between unknown speech and reference templates
    • 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
    • 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
    • G10L15/26—Speech to text systems

Definitions

  • the invention belongs to the field of intelligent terminal applications, and in particular relates to a voice taxiing method and system.
  • the invention is to overcome the defect that it is difficult to determine an accurate taxi order when the address information is not unique in the voice taxiing process in the prior art, and to provide a voice taxiing method and system.
  • a voice taxiing method comprising:
  • the prompting information is used to prompt to acquire the voice supplemental information of the passenger again;
  • a taxi order is generated based on the unique corresponding address information.
  • the determining whether the address information corresponding to the destination information is unique and if not, determining, in the step of generating the prompt information, that the address information corresponding to the destination is the unique correspondence Address information, and then performing the step of generating a taxi order based on the unique corresponding address information.
  • the voice taxiing method before the step of extracting destination information from the voice information, the voice taxiing method further includes:
  • the voice information is the taxi information, and if so, the destination information is extracted from the voice information.
  • the step of determining whether the voice information is the taxi information, and if yes, performing the step of extracting the destination information from the voice information specifically includes:
  • the semantic inventory storing a standard text of taxi
  • the step of extracting destination information from the voice information further includes:
  • the step of generating a taxi order according to the unique corresponding address information generates the taxi order according to the departure place information and the unique corresponding address information.
  • the step of generating the prompt information further includes:
  • Pre-storing an address library where the address library stores a mapping of place name information and geographic location information;
  • Detecting whether the destination information includes the place name information and if so, acquiring geographical location information corresponding to the place name information from the address library, and using the geographical location information corresponding to the place name information as the Uniquely corresponding address information, and then performing the step of generating a taxi order based on the unique corresponding address information; if not, determining whether the address information corresponding to the destination information is unique.
  • the voice taxiing method further includes:
  • the response information is identified. If the response information is a confirmation of a taxi order, the taxi order is sent to the driver terminal; if the response message is a cancellation of the taxi order, the taxi order is cancelled.
  • a voice taxi system comprising a voice information acquisition module, an information extraction module, a destination information judgment module, a prompt module, an address information generation module, and a taxi order generation module;
  • the voice information acquiring module is configured to acquire voice information of a passenger
  • the information extraction module is configured to extract destination information from the voice information
  • the destination information determining module is configured to determine whether the address information corresponding to the destination information is unique, and if not, invoke the prompting module;
  • the prompting module is configured to generate prompt information; the prompt information is used to prompt to acquire the voice supplemental information of the passenger again;
  • the voice information acquiring module is further configured to acquire voice supplemental information of the passenger;
  • the information extraction module is further configured to extract destination supplemental information from the voice supplemental information
  • the address information generating module is configured to determine unique corresponding address information according to the destination information and the destination supplementary information;
  • the taxi order generating module is configured to generate a taxi order according to the unique corresponding address information.
  • the destination information determining module is further configured to: when the determination is yes, confirm that the address information corresponding to the destination is the unique corresponding address information, and then invoke the taxi order generating module.
  • the voice taxi system further includes a taxi information determination module
  • the taxi information determining module is configured to determine whether the voice information is a taxi information, and if yes, invoke the information extraction module to extract the destination information.
  • the taxi information determination module includes a semantic library, a text information conversion module, a calculation module, and a matching degree determination module;
  • the semantic stock stores a standard text of taxis
  • the text information conversion module is configured to convert the voice information into text information
  • the calculating module is configured to compare the text information with the taxi standard text, and calculate a matching degree between the text information and the taxi standard text;
  • the matching degree determining module is configured to determine whether the matching degree is greater than a preset matching threshold, and if yes, confirm that the voice information is taxiing information, and then invoke the information extraction module to extract the destination information.
  • the voice taxi system further includes an information detection module and a departure location information confirmation module;
  • the information detecting module is configured to detect whether the voice information includes the departure location information, and if yes, invoke the information extraction module to extract the departure location information, and if not, invoke the departure location information confirmation module;
  • the departure location information confirmation module is configured to use the current location information of the passenger as the departure location information
  • the taxi order generating module is further configured to generate the taxi order according to the departure place information and the unique corresponding address information.
  • the destination information determining module includes an address library and a geographic location information acquiring module
  • the address store stores a mapping of place name information and geographic location information
  • the information detecting module is further configured to detect whether the destination information includes the place name information, and if yes, invoke the geographic location information obtaining module, and if not, invoke the destination information determining module;
  • the geographic location information obtaining module is configured to obtain the geographical location information corresponding to the geographical name information from the address database, and use the geographical location information corresponding to the geographical name information as the unique corresponding address information.
  • the taxi order generation module is then invoked.
  • the voice taxi system further includes a confirmation request generating module, a response information acquiring module, a response information identifying module, an information sending module, and a cancel order module;
  • the confirmation request generating module is configured to generate a confirmation request of the taxi order
  • the response information obtaining module is configured to obtain response information sent by the passenger according to the confirmation request;
  • the response information identifying module is configured to identify the response information, and if the response information is to confirm a taxi order, the information sending module is invoked, and if the response information is canceling a taxi order, the cancel order module is invoked;
  • the information sending module is configured to send the taxi order to a driver terminal
  • the cancel order module is used to cancel the taxi order.
  • An electronic device includes a memory, a processor, and a computer program stored on the memory and operable on the processor, the processor implementing the voice calling method described above when the computer program is executed.
  • a readable storage medium having stored thereon a computer program that, when executed by a processor, implements the steps of the voice-taking method described above.
  • the positive progress of the invention lies in: extracting the address information in the voice information on the basis of analyzing the voice information of the passenger, and if the address information is not unique, guiding the passenger to input the supplementary voice information, further clarifying the address information, and generating an accurate image. Orders that enhance the passenger's experience of demanding a voice call.
  • Embodiment 1 is a flow chart of a voice taking method according to Embodiment 1 of the present invention.
  • Embodiment 2 is a flow chart of a voice taking method according to Embodiment 2 of the present invention.
  • FIG. 3 is a specific flowchart of step 111 in the voice taking method according to Embodiment 2 of the present invention.
  • FIG. 4 is a specific flowchart of step 120 in the voice taking method according to Embodiment 3 of the present invention.
  • FIG. 5 is a partial flowchart of step 130 in the voice taking method according to Embodiment 4 of the present invention.
  • FIG. 6 is a flowchart of a voice taxiing method according to Embodiment 5 of the present invention.
  • FIG. 7 is a structural block diagram of a voice taxi system according to Embodiment 6 of the present invention.
  • FIG. 8 is a structural block diagram of a voice taxi system according to Embodiment 7 of the present invention.
  • FIG. 9 is a block diagram showing the structure of a taxi information determination module of a voice taxi system according to Embodiment 7 of the present invention.
  • FIG. 10 is a structural block diagram of a voice taxi system according to Embodiment 8 of the present invention.
  • Figure 11 is a block diagram showing the structure of a destination information judging module of a voice taxi system according to Embodiment 9 of the present invention.
  • FIG. 12 is a structural block diagram of a voice taxi system according to Embodiment 10 of the present invention.
  • FIG. 13 is a schematic structural diagram of an electronic device according to Embodiment 11 of the present invention.
  • a voice taxiing method as shown in FIG. 1, the voice taxiing method includes:
  • Step 110 Acquire voice information of a passenger
  • Step 120 Extract destination information from voice information.
  • Step 130 determining whether the address information corresponding to the destination information is unique, and if not, executing step 140;
  • Step 140 Generate prompt information; the prompt information is used to prompt to obtain the passenger's voice supplemental information again;
  • Step 150 Acquire voice supplement information of the passenger
  • Step 160 Extract destination supplemental information from the voice supplemental information
  • Step 170 Determine unique corresponding address information according to the destination information and the destination supplementary information.
  • Step 180 Generate a taxi order according to the unique corresponding address information.
  • the passenger said "Help me call a taxi to Haidilao"
  • the extracted destination information is "Hai Dian”
  • the address information corresponding to "Hai Dian” is multiple.
  • the prompt message "has been queried.”
  • the passengers say “I am going to Haidilao near Sihui Bridge” to extract the destination according to the passenger's supplementary voice information.
  • the supplementary information is “Sihui Bridge”, and the unique corresponding address information is determined according to the destination information and the supplementary destination information as “Haidilao Hot Pot Red Temple Store”, and a taxi is generated according to the unique corresponding address information “Haidilao Hot Pot Red Temple Store”.
  • the “Yizhuang Hualian Store” corresponding to the “confirmation” of the supplementary information determines the unique corresponding address information according to the destination information and the supplementary destination information as “Haidilao Yizhuang Hualian Store”, according to the unique corresponding address information “Haidilao Yizhuanghua”
  • the joint store "generates a taxi order.
  • the voice taxiing method further includes:
  • step 130 determines whether the determination in step 130 is YES, the address information corresponding to the destination is the only corresponding address information, and then step 180 is performed;
  • the passenger said, “Help me call a taxi to the Haidilao Hongmiao store near Sihui Bridge”.
  • the destination information extracted is “Haidilao Hot Pot Red Temple Store”, and the address information corresponding to the destination address information is unique. Then, directly generate a taxi order.
  • the address information in the voice information is extracted, and if the address information is not unique, the passenger is guided to input supplementary voice information, further clarify the address information, generate an accurate order, and improve The experience of passengers passing the demand for a taxi.
  • the voice taxiing method in this embodiment is further improved on the basis of Embodiment 1.
  • the voice taxiing method before step 120 further includes:
  • Step 111 Determine whether the voice information is the taxi information. If yes, execute step 120. If not, confirm that the voice information is not the taxi information.
  • step 111 includes:
  • Step 111a preset a semantic library, and the semantic inventory stores a standard text of taxis
  • Step 111b Convert voice information into text information
  • Step 111c comparing the text information with the taxi standard text, and calculating a matching degree between the text information and the taxi standard text;
  • Step 111d Determine whether the matching degree is greater than a preset matching threshold. If yes, confirm that the voice information is the taxi information, and then perform step 120. If not, confirm that the voice information is not the taxi information.
  • the voice taking process uses multiple rounds of conversation, and each session is processed in a fuzzy manner.
  • the fuzzy mode can use fuzzy mathematics for fuzzy quantization, and the real number between 0 and 1 is used to represent the degree of membership.
  • the value of the subordinate function of the standard word and phrase satisfying the predetermined standard can be set to 1, then other similar meanings and words that can express similar thoughts (semantics) can be used for 0 to 1 A continuous number between them to characterize the degree to which it belongs to the "correct sentence.”
  • the statement information is compared with the taxi standard text pre-stored in the semantic library, and if the matching degree of the comparison satisfies the predetermined requirement (for example, the matching degree is above 0.7), the ratio is Passed.
  • the voice information of the passenger is analyzed in advance, and when the voice information of the passenger is determined to be the taxi information, the taxi order is further generated based on the voice information.
  • the step 120 further includes:
  • Step 1201 detecting whether the voice information includes the departure location information, and if so, executing step 1202, if not, executing step 1203;
  • Step 1202 Extract the departure place information
  • Step 1203 Taking the current location information of the passenger as the departure location information
  • step 180 a taxi order is generated based on the departure location information and the unique corresponding address information.
  • the generation of the taxi order requires not only the destination information but also the departure location information.
  • the departure location information is extracted from the voice information, and the address information corresponding to the departure location information is obtained. If it is not unique, the unique address information corresponding to the departure location information may be obtained by referring to the method for obtaining the unique address information; if the location information is not included in the voice information sent by the passenger, the current location information of the passenger is used as the departure location information. .
  • the step 130 further includes:
  • Step 1301 Pre-storing an address library, where the address store stores mapping of place name information and geographic location information;
  • Step 1302 Detect whether the destination information includes the place name information, and if yes, execute step 1303, and if not, determine whether the address information corresponding to the destination information is unique;
  • step 1303 the geographic location information corresponding to the geographical name information is obtained from the address database, and the geographical location information corresponding to the geographical name information is used as the unique corresponding address information, and then step 180 is performed.
  • the passenger can pre-set common location information, such as "home", "company” and the like.
  • the voice information of the passenger includes the above-mentioned common location information
  • the geographical location information corresponding to the geographical name information is directly obtained from the address database, and the geographical location information corresponding to the geographical name information is used as the unique corresponding address information, which needs to be described.
  • the departure point information in the voice information includes the above-mentioned place name
  • the unique corresponding departure place information can also be obtained by the above method.
  • the voice taxiing method of this embodiment is further improved on the basis of the embodiment 1. As shown in FIG. 6, the voice taxiing method further includes:
  • Step 190 Generate a confirmation request for a taxi order
  • Step 1100 Acquire response information sent by the passenger according to the confirmation request
  • step 1110 the response information is identified. If the response message is to confirm the taxi order, step 1110a is performed. If the response message is to cancel the taxi order, step 1110b is performed.
  • Step 1110a sending a taxi order to the driver terminal
  • Step 1110b cancel the taxi order.
  • a confirmation request is generated and sent to the passenger for further confirmation. If the passenger replies with an utterance of canceling the order intention, such as “cancel”, the automatic recognition and answer: “order cancelled” (or similar utterance) And the service is terminated. If the passenger answers the words “confirm” and other means of confirming the order's intention, it will automatically recognize and answer: "order confirmed” (or similar discourse) and send a taxi order to the driver.
  • a voice taxi system as shown in FIG. 7, the voice taxi system includes a voice information acquisition module 1, an information extraction module 2, a destination information determination module 3, a prompt module 4, an address information generation module 5, and a taxi order generation Module 6;
  • the voice information acquiring module 1 is configured to acquire voice information of a passenger
  • the information extraction module 2 is configured to extract destination information from the voice information.
  • the destination information judging module 3 is configured to determine whether the address information corresponding to the destination information is unique, and if not, invoke the prompting module 4;
  • the prompting module 4 is configured to generate prompt information; the prompt information is used to prompt to acquire the voice supplemental information of the passenger again;
  • the voice information acquiring module 1 is further configured to acquire voice supplemental information of the passenger;
  • the information extraction module 2 is further configured to extract destination supplemental information from the voice supplemental information
  • the address information generating module 5 is configured to determine unique corresponding address information according to the destination information and the destination supplementary information;
  • the taxi order generating module 6 is configured to generate a taxi order according to the unique corresponding address information.
  • the destination information determining module 3 is further configured to: when the determination is yes, confirm that the address information corresponding to the destination is the unique corresponding address information, and then invoke the taxi order generating module 6.
  • the passenger said “Help me call a taxi to Haidilao”
  • the destination information extracted by the information extraction module is "Hai Dian”
  • the destination information judgment module judges that the address information corresponding to "Hai Dian” is plural.
  • the prompt module generates a prompt message "After multiple locations have been queried, you can ask me about the sea fishing near a certain place, such as the sea fishing at the Xizhimen annex.” The passenger said “I am going to the sea fishing near Sihui Bridge”.
  • the information extraction module extracts the destination supplementary information as “Sihui Bridge”, and the taxi order generating module determines the unique corresponding address information according to the destination information and the supplementary destination information as “Haidilao Hot Pot Red Temple Store” ", according to the unique address information "Haidilao Hot Pot Red Temple Store” to generate a taxi order;
  • the prompt module generates the prompt message "The nearest to you is Haidilao Yizhuang Hualian Store, do you confirm Haidilao Yizhuang Hualian Store as the destination address?", the passengers say “confirmation address", according to the passenger's supplementary voice information, information
  • the extraction module extracts the "Yizhuang Hualian Store” corresponding to the destination supplementary information "confirmation”, and the taxi order generating module determines the unique corresponding address information according to the destination information and the supplementary destination information as "Haidilao Yizhuang Hualian Store", according to The only corresponding address information "Haidilao Yizhuang Hualian Store” generates a taxi order.
  • the address information in the voice information is extracted, and if the address information is not unique, the passenger is guided to input supplementary voice information, further clarify the address information, generate an accurate order, and improve The experience of passengers passing the demand for a taxi.
  • the voice-taking system of the present embodiment is further improved on the basis of the embodiment 6, the voice-taking system further includes a taxi information determining module 7;
  • the taxi information determining module 7 is configured to determine whether the voice information is a taxi information, and if so, invoke the information extraction module 2 to extract the destination information.
  • the taxi information determination module 7 includes a semantic library 71, a text information conversion module 72, a calculation module 73, and a matching degree determination module 74;
  • the semantic library 71 stores a taxi standard text
  • the text information conversion module 72 is configured to convert the voice information into text information
  • the calculating module 73 is configured to compare the text information with the taxi standard text, and calculate a matching degree between the text information and the taxi standard text;
  • the matching degree determining module 74 is configured to determine whether the matching degree is greater than a preset matching threshold, and if yes, confirm that the voice information is taxiing information, and then invoke the information extracting module 2 to extract the destination information.
  • the voice taking process uses multiple rounds of conversation, and each session is processed in a fuzzy manner.
  • the fuzzy mode can use fuzzy mathematics for fuzzy quantization, and the real number between 0 and 1 is used to represent the degree of membership.
  • the value of the subordinate function of the standard word and phrase satisfying the predetermined standard can be set to 1, then other similar meanings and words that can express similar thoughts (semantics) can be used for 0 to 1 A continuous number between them to characterize the degree to which it belongs to the "correct sentence.”
  • the statement information is compared with the taxi standard text pre-stored in the semantic library, and if the matching degree of the comparison satisfies the predetermined requirement (for example, the matching degree is above 0.7), the ratio is Passed.
  • the voice information of the passenger is analyzed in advance, and when the voice information of the passenger is determined to be the taxi information, the taxi order is further generated based on the voice information.
  • the voice-taking system of the present embodiment is further improved on the basis of the embodiment 6, as shown in FIG. 10, the voice-taking system further includes an information detecting module 8 and a departure place information confirming module 9;
  • the information detecting module 8 is configured to detect whether the voice information includes the departure location information, and if so, the information extraction module 2 is invoked to extract the departure location information, and if not, the departure location information confirmation module 9 is invoked;
  • the departure location information confirmation module 9 is configured to use the current location information of the passenger as the departure location information
  • the taxi order generating module 6 is further configured to generate the taxi order according to the departure place information and the unique corresponding address information.
  • the generating the taxi order requires not only the destination information but also the departure location information.
  • the information extraction module extracts the departure location information from the voice information, and if the departure location information corresponds to The address information is not unique, and the unique address information corresponding to the departure location information may be obtained by referring to the method for obtaining the unique address information; if the voice information sent by the passenger does not include the departure location information, the current location information of the passenger is taken as its Departure information.
  • the destination information determining module 3 includes an address library 31 and a geographic location information acquiring module 32;
  • the address library 31 stores a mapping of place name information and geographic location information
  • the information detecting module 11 is further configured to detect whether the destination information includes the place name information, and if so, the geographic location information obtaining module 32 is invoked, and if not, the destination information determining module 3 is invoked;
  • the geographic location information obtaining module 32 is configured to obtain geographic location information corresponding to the geographical name information from the address pool 31, and use the geographical location information corresponding to the geographical location information as the unique corresponding address. The information is then invoked by the taxi order generating module 6.
  • the passenger can pre-set common location information, such as "home", "company” and the like.
  • the voice information of the passenger includes the above-mentioned common location information
  • the geographical location information corresponding to the geographical name information is directly obtained from the address database, and the geographical location information corresponding to the geographical name information is used as the unique corresponding address information, which needs to be described.
  • the departure point information in the voice information includes the above-mentioned place name
  • the unique corresponding departure place information can also be obtained by the above method.
  • the voice-taking system of the present embodiment is further improved on the basis of the embodiment 6. Specifically, as shown in FIG. 12, the voice-taking system further includes a confirmation request generating module 10, a response information acquiring module 11, and a response information identifying module 12, Information sending module 13 and cancel order module 14;
  • the confirmation request generating module 10 is configured to generate a confirmation request of the taxi order
  • the response information obtaining module 11 is configured to acquire response information sent by the passenger according to the confirmation request;
  • the response information identifying module 12 is configured to identify the response information, and if the response information is to confirm a taxi order, the information sending module 13 is invoked, and if the response information is canceling a taxi order, the canceling order is invoked.
  • the information sending module 13 is configured to send the taxi order to the driver terminal
  • the cancel order module 14 is used to cancel the taxi order.
  • a confirmation request is generated and sent to the passenger for further confirmation. If the passenger replies with an utterance of canceling the order intention, such as “cancel”, the automatic recognition and answer: “order cancelled” (or similar utterance) And the service is terminated. If the passenger answers the words “confirm” and other means of confirming the order's intention, it will automatically recognize and answer: "order confirmed” (or similar discourse) and send a taxi order to the driver.
  • An electronic device comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, the processor executing the computer program to implement any one of the embodiments 1-5 Voice taxi method.
  • the electronic device may be in the form of a general-purpose computing device, for example, it may be a server device or may be in the form of a smart mobile terminal, for example, it may be a mobile phone, a tablet, or the like.
  • FIG. 13 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure.
  • FIG. 13 illustrates a block diagram of an exemplary electronic device 90 suitable for use in implementing embodiments of the present invention.
  • the electronic device 90 shown in FIG. 13 is merely an example and should not impose any limitation on the function and scope of use of the embodiments of the present invention.
  • components of electronic device 90 may include, but are not limited to, at least one processor 91, at least one memory 92, and a bus 93 that connects different system components, including memory 92 and processor 91.
  • the bus 93 includes a data bus, an address bus, and a control bus.
  • Memory 92 may include volatile memory, such as random access memory (RAM) 921 and/or cache memory 922, and may further include read only memory (ROM) 923.
  • RAM random access memory
  • ROM read only memory
  • the memory 92 can also include a program tool 925 having a set (at least one) of program modules 924, including but not limited to: an operating system, one or more applications, other program modules, and program data, in these examples. Each or some combination may include an implementation of a network environment.
  • the processor 91 executes various functional applications and data processing by running a computer program stored in the memory 92.
  • Electronic device 90 may also be in communication with one or more external devices 94 (eg, a keyboard, pointing device, etc.). This communication can take place via an input/output (I/O) interface 95. Also, electronic device 90 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) via network adapter 96. Network adapter 96 communicates with other modules of electronic device 90 via bus 93. It should be understood that although not shown in the figures, other hardware and/or software modules may be utilized in conjunction with electronic device 90, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk arrays). Systems, tape drives, and data backup storage systems.
  • a computer readable storage medium having stored thereon a computer program, the program being executed by a processor to implement the steps of the voice caring method of any of embodiments 1-5.
  • the readable storage medium may be more specifically included, but not limited to: a portable disk, a hard disk, a random access memory, a read only memory, a wipeable programmable read only memory, an optical storage device, a magnetic storage device, or any of the above. A suitable combination.
  • the present invention may also be embodied in the form of a program product comprising program code for causing the terminal device to perform implementation when the program product is run on a terminal device.
  • program code for causing the terminal device to perform implementation when the program product is run on a terminal device.
  • program code for executing the invention may be written in any combination of one or more programming languages, the program code being executable entirely on the user device, partly on the user device, as an independent
  • the software package is executed, partially on the user device, partially on the remote device, or entirely on the remote device.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Business, Economics & Management (AREA)
  • Multimedia (AREA)
  • Acoustics & Sound (AREA)
  • Human Computer Interaction (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Traffic Control Systems (AREA)
  • Telephonic Communication Services (AREA)
  • Navigation (AREA)

Abstract

本发明公开了一种语音打车方法及***，所述语音打车方法包括：获取乘客的语音信息；从语音信息中提取目的地信息；判断与目的地信息对应的地址信息是否唯一，若否，生成提示信息；提示信息用于提示再次获取乘客的语音补充信息；获取乘客的语音补充信息；从语音补充信息中提取目的地补充信息；根据目的地信息和目的地补充信息确定唯一对应的地址信息；根据唯一对应的地址信息生成打车订单。本发明在对乘客的语音信息进行分析的基础上，当从语音信息中提取出的地址信息不唯一时，引导乘客输入补充语音信息，进一步明确地址信息，生成准确的订单，提高乘客通过语音打车的需求的体验。

Description

语音打车方法及***
本申请要求申请日为2017年12月20日的中国专利申请CN201711384195.4的优先权。本申请引用上述中国专利申请的全文。
技术领域
本发明属于智能终端应用领域，特别涉及一种语音打车方法及***。
背景技术
随着互联网的发展，打车出行作为现代人的刚需，流程也越来越简化。市场上不仅有出租车公司统一调度的电话约车***，也出现了基于手持智能终端的打车软件。人们现在已经习惯于使用网络来使用打车服务，只需要在乘客端输入出发地点和目的地点，服务器生成订单并向符合条件的司机端发出推送通知，司机根据自身的运营情况选择接单就可以让乘客和司机完成打车交易。但是，对于乘客来说，需要手动输入起始点和目的地，依赖手工填写来完成，不够便捷；对于司机而言，则需要再车载移动终端上人工手动确认订单，对于驾驶中的司机来说是非常不便利且比较危险的。
基于上述问题，现有的技术中，对语音打车进行了一些尝试，但是并没有让用户确认的机制，尤其是ASR(Automatic Speech Recognition，自动语音识别)准确率不能达到100％的技术背景下，识别错误势必会导致乘客和司机的困扰，除此之外，当乘客发出的打车请求中，所述的目的地信息结果如果不唯一，例如乘客说“我要去海底捞”，现有技术的解决方案中只是将地理位置信息打包发给司机端，不能将准确信息筛选出来，因此在实际使用时，体验难以得到保证。
发明内容
本发明是为了克服现有技术中语音打车过程中地址信息不唯一时难以确定准确的打车订单的缺陷，提供一种语音打车方法及***。
本发明通过以下技术方案来解决上述问题：
一种语音打车方法，包括：
获取乘客的语音信息；
从所述语音信息中提取目的地信息；
判断与所述目的地信息对应的地址信息是否唯一，若否，则生成提示信息；所述提 示信息用于提示再次获取所述乘客的语音补充信息；
获取所述乘客的语音补充信息；
从所述语音补充信息中提取目的地补充信息；
根据所述目的地信息和所述目的地补充信息确定唯一对应的地址信息；
根据所述唯一对应的地址信息生成打车订单。
较佳地，所述判断与所述目的地信息对应的地址信息是否唯一，若否，则生成提示信息的步骤中若判断为是，则确认所述目的地对应的地址信息为所述唯一对应的地址信息，然后执行所述根据所述唯一对应的地址信息生成打车订单的步骤。
较佳地，所述从所述语音信息中提取目的地信息的步骤之前，所述语音打车方法还包括：
判断所述语音信息是否为打车信息，若是，则执行从所述语音信息中提取目的地信息。
较佳地，所述判断所述语音信息是否为打车信息，若是，则执行从所述语音信息中提取目的地信息的步骤具体包括：
预设一语义库，所述语义库存储有打车标准文本；
将所述语音信息转化为文本信息；
将所述文本信息与所述打车标准文本进行比对，计算所述文本信息与所述打车标准文本的匹配度；
判断所述匹配度是否大于预设的匹配阈值，若是，则确认所述语音信息为打车信息，然后执行从所述语音信息中提取目的地信息。
较佳地，所述从所述语音信息中提取目的地信息的步骤还包括：
检测所述语音信息是否包括出发地信息，若是，则提取所述出发地信息，若否，则将所述乘客的当前位置信息作为所述出发地信息；
所述根据所述唯一对应的地址信息生成打车订单的步骤中根据所述出发地信息和所述唯一对应的地址信息生成所述打车订单。
较佳地，所述判断与所述目的地信息对应的地址信息是否唯一，若否，则生成提示信息的步骤还包括：
预存一地址库，所述地址库存储有地名信息与地理位置信息的映射；
检测所述目的地信息是否包括所述地名信息，若是，从所述地址库中获取与所述地名信息对应的地理位置信息，并将所述与所述地名信息对应的地理位置信息作为所述唯一对应的地址信息，然后执行所述根据所述唯一对应的地址信息生成打车订单的步骤； 若否，则判断与所述目的地信息对应的地址信息是否唯一。
较佳地，所述语音打车方法还包括：
生成所述打车订单的确认请求；
获取乘客根据所述确认请求发出的响应信息；
识别所述响应信息，若所述响应信息为确认打车订单，则将打车订单发送至司机端；若所述响应信息为取消打车订单，则取消所述打车订单。
一种语音打车***，包括语音信息获取模块、信息提取模块、目的地信息判断模块、提示模块、地址信息生成模块和打车订单生成模块；
所述语音信息获取模块用于获取乘客的语音信息；
所述信息提取模块用于从所述语音信息中提取目的地信息；
所述目的地信息判断模块用于判断与所述目的地信息对应的地址信息是否唯一，若否，调用所述提示模块；
所述提示模块用于生成提示信息；所述提示信息用于提示再次获取所述乘客的语音补充信息；
所述语音信息获取模块还用于获取所述乘客的语音补充信息；
所述信息提取模块还用于从所述语音补充信息中提取目的地补充信息；
所述地址信息生成模块用于根据所述目的地信息和所述目的地补充信息确定唯一对应的地址信息；
所述打车订单生成模块用于根据所述唯一对应的地址信息生成打车订单。
较佳地，所述目的地信息判断模块还用于在判断为是时，确认所述目的地对应的地址信息为所述唯一对应的地址信息，然后调用所述打车订单生成模块。
较佳地，所述语音打车***还包括打车信息判断模块；
所述打车信息判断模块用于判断所述语音信息是否为打车信息，若是，则调用所述信息提取模块提取所述目的地信息。
较佳地，所述打车信息判断模块包括语义库、文本信息转化模块、计算模块和匹配度判断模块；
所述语义库存储有打车标准文本；
所述文本信息转化模块用于将所述语音信息转化为文本信息；
所述计算模块用于将所述文本信息与所述打车标准文本进行比对，计算所述文本信息与所述打车标准文本的匹配度；
所述匹配度判断模块用于判断所述匹配度是否大于预设的匹配阈值，若是，则确认 所述语音信息为打车信息，然后调用所述信息提取模块提取所述目的地信息。
较佳地，所述语音打车***还包括信息检测模块和出发地信息确认模块；
所述信息检测模块用于检测所述语音信息是否包括出发地信息，若是，则调用所述信息提取模块提取所述出发地信息，若否，则调用所述出发地信息确认模块；
所述出发地信息确认模块用于将所述乘客的当前位置信息作为所述出发地信息；
所述打车订单生成模块还用于根据所述出发地信息和所述唯一对应的地址信息生成所述打车订单。
较佳地，所述目的地信息判断模块包括地址库和地理位置信息获取模块；
所述地址库存储有地名信息与地理位置信息的映射；
所述信息检测模块还用于检测所述目的地信息是否包括所述地名信息，若是，则调用所述地理位置信息获取模块，若否，则调用所述目的地信息判断模块；
所述地理位置信息获取模块用于从所述地址库中获取与所述地名信息对应的地理位置信息，并将所述与所述地名信息对应的地理位置信息作为所述唯一对应的地址信息，然后调用所述打车订单生成模块。
较佳地，所述语音打车***还包括确认请求生成模块、响应信息获取模块、响应信息识别模块、信息发送模块和取消订单模块；
所述确认请求生成模块用于生成所述打车订单的确认请求；
所述响应信息获取模块用于获取乘客根据所述确认请求发出的响应信息；
所述响应信息识别模块用于识别所述响应信息，若所述响应信息为确认打车订单，则调用所述信息发送模块，若所述响应信息为取消打车订单，则调用所述取消订单模块；
所述信息发送模块用于将所述打车订单发送至司机端；
所述取消订单模块用于取消所述打车订单。
一种电子设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述的语音打车方法。
一种可读存储介质，其上存储有计算机程序，所述程序被处理器执行时实现上述的语音打车方法的步骤。
本发明的积极进步效果在于：在对乘客的语音信息进行分析的基础上，提取出语音信息中的地址信息，若地址信息不唯一，引导乘客输入补充语音信息，进一步明确地址信息，生成准确的订单，提高乘客通过语音打车的需求的体验。
附图说明
图1为本发明实施例1的语音打车方法的流程图。
图2为本发明实施例2的语音打车方法的流程图。
图3为本发明实施例2的语音打车方法中步骤111的具体流程图。
图4为本发明实施例3的语音打车方法中步骤120的具体流程图。
图5为本发明实施例4的语音打车方法中步骤130的部分流程图。
图6为本发明实施例5的语音打车方法的流程图。
图7为本发明实施例6的语音打车***的结构框图。
图8为本发明实施例7的语音打车***的结构框图。
图9为本发明实施例7的语音打车***的打车信息判断模块的结构框图。
图10为本发明实施例8的语音打车***的结构框图。
图11为本发明实施例9的语音打车***的目的地信息判断模块的结构框图。
图12为本发明实施例10的语音打车***的结构框图。
图13为本发明实施例11的电子设备的结构示意图。
具体实施方式
下面通过实施例的方式进一步说明本发明，但并不因此将本发明限制在所述的实施例范围之中。
实施例1
一种语音打车方法，具体如图1所示，语音打车方法包括：
步骤110、获取乘客的语音信息；
步骤120、从语音信息中提取目的地信息；
步骤130、判断与目的地信息对应的地址信息是否唯一，若否，执行步骤140；
步骤140、生成提示信息；提示信息用于提示再次获取乘客的语音补充信息；
步骤150、获取乘客的语音补充信息；
步骤160、从语音补充信息中提取目的地补充信息；
步骤170、根据目的地信息和目的地补充信息确定唯一对应的地址信息；
步骤180、根据唯一对应的地址信息生成打车订单。
比如乘客说出“帮我叫个出租车去海底捞”，提取出的目的地信息为“海底捞”，与“海底捞”对应的地址信息为多个，此时生成提示信息“已查询到多个地点，你可以问我某地附近的海底捞，比如西直门附件的海底捞”，乘客说出“我要去四惠桥附近的海底捞”，根据乘客的补充语音信息，提取出目的地补充信息为“四惠桥”，根据目的地信息 和补充目的地信息确定唯一对应的地址信息为“海底捞火锅红庙店”，根据唯一对应的地址信息“海底捞火锅红庙店”生成打车订单；
或者提示信息“距离您最近的是海底捞亦庄华联店，是否确认海底捞亦庄华联店为目的地地址？”，乘客说出“确认地址”，根据乘客的补充语音信息，提取出目的地补充信息“确认”对应的“亦庄华联店”，根据目的地信息和补充目的地信息确定唯一对应的地址信息为“海底捞亦庄华联店”，根据唯一对应的地址信息“海底捞亦庄华联店”生成打车订单。
所述语音打车方法还包括：
步骤130中若判断为是，则确认目的地对应的地址信息为唯一对应的地址信息，然后执行步骤180；
比如乘客说出“帮我叫个出租车去四惠桥附近的海底捞红庙店”，提取的目的地信息为“海底捞火锅红庙店”，该目的地地址信息对应的地址信息为唯一的，那么直接生成打车订单。
本实施例中，在对乘客的语音信息进行分析的基础上，提取出语音信息中的地址信息，若地址信息不唯一，引导乘客输入补充语音信息，进一步明确地址信息，生成准确的订单，提高乘客通过语音打车的需求的体验。
实施例2
如图2-3所示，本实施例的语音打车方法是在实施例1的基础上进一步改进，步骤120之前语音打车方法还包括：
步骤111、判断语音信息是否为打车信息，若是，则执行步骤120，若否，确认该语音信息不是打车信息。
具体的，步骤111包括：
步骤111a、预设一语义库，语义库存储有打车标准文本；
步骤111b、将语音信息转化为文本信息；
步骤111c、将文本信息与打车标准文本进行比对，计算文本信息与打车标准文本的匹配度；
步骤111d、判断匹配度是否大于预设的匹配阈值，若是，则确认语音信息为打车信息，然后执行步骤120，若否，确认该语音信息不是打车信息。
本实施例中，语音打车过程均采用多轮会话，并且采用模糊方式处理每段对话，模糊方式可以采用模糊数学进行模糊量化，用介于0和1之间的实数来表示隶属程度。相应地，在语义识别上，可以将满足预定标准的标准词句的从属函数值定为1，那么，其他 近义的，以及能表达相仿的思想(语义)的词句，就可以用以0到1之间的连续数来表征它从属于“正确句子”的隶属程度。这样，获取到的乘客的语音信息后，将该语句信息与预存于语义库中的打车标准文本进行比对，如果比对的匹配度满足预定要求(如匹配度在0.7以上)，则表示比对通过。本实施例中预先对乘客的语音信息进行分析，在判断乘客的语音信息为打车信息时，才进一步根据该语音信息生成打车订单。
实施例3
本实施例的语音打车方法是在实施例1的基础上进一步改进，具体如图4所示，步骤120还包括：
步骤1201、检测语音信息是否包括出发地信息，若是，则执行步骤1202，若否，则执行步骤1203；
步骤1202、提取出发地信息；
步骤1203、将乘客的当前位置信息作为出发地信息；
步骤180中根据出发地信息和唯一对应的地址信息生成打车订单。
本实施例中，生成打车订单不仅需要目的地信息，还需要出发地信息，当乘客发出的语音信息中包含出发地信息时，从语音信息中提取出发地信息，若出发地信息对应的地址信息不唯一，可以参照获取唯一的地址信息的方式获取与出发地信息相对应的唯一的地址信息；若乘客发出的语音信息中不包含出发地信息，则将乘客的当前位置信息作为其出发地信息。
实施例4
本实施例的语音打车方法是在实施例1的基础上进一步改进，具体如图5所示，步骤130还包括：
步骤1301、预存一地址库，地址库存储有地名信息与地理位置信息的映射；
步骤1302、检测目的地信息是否包括地名信息，若是，执行步骤1303，若否，则判断与目的地信息对应的地址信息是否唯一；
步骤1303、从地址库中获取与地名信息对应的地理位置信息，并将与地名信息对应的地理位置信息作为唯一对应的地址信息，然后执行步骤180。
本实施例中，乘客可以预设置常用地点信息，如“家”、“公司”等信息。当乘客的语音信息包含上述常用地点信息时，直接从地址库中获取与地名信息对应的地理位置信息，并将与地名信息对应的地理位置信息作为唯一对应的地址信息，需要说明的是，此处若语音信息中的出发点信息包含上述地名时，同样也能够通过上述方法得知唯一对应的出发地信息。
实施例5
本实施例的语音打车方法是在实施例1的基础上进一步改进，具体如图6所示，语音打车方法还包括：
步骤190、生成打车订单的确认请求；
步骤1100、获取乘客根据确认请求发出的响应信息；
步骤1110、识别响应信息，若响应信息为确认打车订单，则执行步骤1110a，若响应信息为取消打车订单，则执行步骤1110b。
步骤1110a、将打车订单发送至司机端；
步骤1110b、取消打车订单。
本实施例中，打车订单生成后需要生成确认请求发给乘客进一步确认，若乘客回答：“取消”等有取消订单意图的话语时，将自动识别并回答：“订单已取消”(或类似话语)并结束此次服务，若乘客回答“确认”等有确认订单意图的话语时，将自动识别并回答：“订单已确认”(或类似话语)并将打车订单发送至司机端。
实施例6
一种语音打车***，具体如图7所示，所述语音打车***包括语音信息获取模块1、信息提取模块2、目的地信息判断模块3、提示模块4、地址信息生成模块5和打车订单生成模块6；
所述语音信息获取模块1用于获取乘客的语音信息；
所述信息提取模块2用于从所述语音信息中提取目的地信息；
所述目的地信息判断模块3用于判断与所述目的地信息对应的地址信息是否唯一，若否，调用所述提示模块4；
所述提示模块4用于生成提示信息；所述提示信息用于提示再次获取所述乘客的语音补充信息；
所述语音信息获取模块1还用于获取所述乘客的语音补充信息；
所述信息提取模块2还用于从所述语音补充信息中提取目的地补充信息；
所述地址信息生成模块5用于根据所述目的地信息和所述目的地补充信息确定唯一对应的地址信息；
所述打车订单生成模块6用于根据所述唯一对应的地址信息生成打车订单。
所述目的地信息判断模块3还用于在判断为是时，确认所述目的地对应的地址信息为所述唯一对应的地址信息，然后调用所述打车订单生成模块6。
比如乘客说出“帮我叫个出租车去海底捞”，信息提取模块提取出的目的地信息为“海 底捞”，目的地信息判断模块判断与“海底捞”对应的地址信息为多个，此时提示模块生成提示信息“已查询到多个地点，你可以问我某地附近的海底捞，比如西直门附件的海底捞”，乘客说出“我要去四惠桥附近的海底捞”，根据乘客的补充语音信息，信息提取模块提取出目的地补充信息为“四惠桥”，打车订单生成模块根据目的地信息和补充目的地信息确定唯一对应的地址信息为“海底捞火锅红庙店”，根据唯一对应的地址信息“海底捞火锅红庙店”生成打车订单；
或者提示模块生成提示信息“距离您最近的是海底捞亦庄华联店，是否确认海底捞亦庄华联店为目的地地址？”，乘客说出“确认地址”，根据乘客的补充语音信息，信息提取模块提取出目的地补充信息“确认”对应的“亦庄华联店”，打车订单生成模块根据目的地信息和补充目的地信息确定唯一对应的地址信息为“海底捞亦庄华联店”，根据唯一对应的地址信息“海底捞亦庄华联店”生成打车订单。
本实施例中，在对乘客的语音信息进行分析的基础上，提取出语音信息中的地址信息，若地址信息不唯一，引导乘客输入补充语音信息，进一步明确地址信息，生成准确的订单，提高乘客通过语音打车的需求的体验。
实施例7
如图8-9所示，本实施例的语音打车***是在实施例6的基础上进一步改进，所述语音打车***还包括打车信息判断模块7；
所述打车信息判断模块7用于判断所述语音信息是否为打车信息，若是，则调用所述信息提取模块2提取所述目的地信息。
具体的，所述打车信息判断模块7包括语义库71、文本信息转化模块72、计算模块73和匹配度判断模块74；
所述语义库71存储有打车标准文本；
所述文本信息转化模块72用于将所述语音信息转化为文本信息；
所述计算模块73用于将所述文本信息与所述打车标准文本进行比对，计算所述文本信息与所述打车标准文本的匹配度；
所述匹配度判断模块74用于判断所述匹配度是否大于预设的匹配阈值，若是，则确认所述语音信息为打车信息，然后调用所述信息提取模块2提取所述目的地信息。
本实施例中，语音打车过程均采用多轮会话，并且采用模糊方式处理每段对话，模糊方式可以采用模糊数学进行模糊量化，用介于0和1之间的实数来表示隶属程度。相应地，在语义识别上，可以将满足预定标准的标准词句的从属函数值定为1，那么，其他近义的，以及能表达相仿的思想(语义)的词句，就可以用以0到1之间的连续数来表 征它从属于“正确句子”的隶属程度。这样，获取到的乘客的语音信息后，将该语句信息与预存于语义库中的打车标准文本进行比对，如果比对的匹配度满足预定要求(如匹配度在0.7以上)，则表示比对通过。本实施例中预先对乘客的语音信息进行分析，在判断乘客的语音信息为打车信息时，才进一步根据该语音信息生成打车订单。
实施例8
本实施例的语音打车***是在实施例6的基础上进一步改进，具体如图10所示，所述语音打车***还包括信息检测模块8和出发地信息确认模块9；
所述信息检测模块8用于检测所述语音信息是否包括出发地信息，若是，则调用所述信息提取模块2提取所述出发地信息，若否，则调用所述出发地信息确认模块9；
所述出发地信息确认模块9用于将所述乘客的当前位置信息作为所述出发地信息；
所述打车订单生成模块6还用于根据所述出发地信息和所述唯一对应的地址信息生成所述打车订单。
本实施例中，生成打车订单不仅需要目的地信息，还需要出发地信息，当乘客发出的语音信息中包含出发地信息时，信息提取模块从语音信息中提取出发地信息，若出发地信息对应的地址信息不唯一，可以参照获取唯一的地址信息的方式获取与出发地信息相对应的唯一的地址信息；若乘客发出的语音信息中不包含出发地信息，则将乘客的当前位置信息作为其出发地信息。
实施例9
本实施例的语音打车***是在实施例8的基础上进一步改进，具体如图11所示，所述目的地信息判断模块3包括地址库31和地理位置信息获取模块32；
所述地址库31存储有地名信息与地理位置信息的映射；
所述信息检测模块11还用于检测所述目的地信息是否包括所述地名信息，若是，则调用所述地理位置信息获取模块32，若否，则调用所述目的地信息判断模块3；
所述地理位置信息获取模块32用于从所述地址库31中获取与所述地名信息对应的地理位置信息，并将所述与所述地名信息对应的地理位置信息作为所述唯一对应的地址信息，然后调用所述打车订单生成模块6。
本实施例中，乘客可以预设置常用地点信息，如“家”、“公司”等信息。当乘客的语音信息包含上述常用地点信息时，直接从地址库中获取与地名信息对应的地理位置信息，并将与地名信息对应的地理位置信息作为唯一对应的地址信息，需要说明的是，此处若语音信息中的出发点信息包含上述地名时，同样也能够通过上述方法得知唯一对应的出发地信息。
实施例10
本实施例的语音打车***是在实施例6的基础上进一步改进，具体如图12所示，所述语音打车***还包括确认请求生成模块10、响应信息获取模块11、响应信息识别模块12、信息发送模块13和取消订单模块14；
所述确认请求生成模块10用于生成所述打车订单的确认请求；
所述响应信息获取模块11用于获取乘客根据所述确认请求发出的响应信息；
所述响应信息识别模块12用于识别所述响应信息，若所述响应信息为确认打车订单，则调用所述信息发送模块13，若所述响应信息为取消打车订单，则调用所述取消订单模块14；
所述信息发送模块13用于将所述打车订单发送至司机端；
所述取消订单模块14用于取消所述打车订单。
本实施例中，打车订单生成后需要生成确认请求发给乘客进一步确认，若乘客回答：“取消”等有取消订单意图的话语时，将自动识别并回答：“订单已取消”(或类似话语)并结束此次服务，若乘客回答“确认”等有确认订单意图的话语时，将自动识别并回答：“订单已确认”(或类似话语)并将打车订单发送至司机端。
实施例11
一种电子设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现实施例1-5中任意一个实施例所述的语音打车方法。
电子设备可以以通用计算设备的形式表现，例如其可以为服务器设备，也可以以智能移动终端的形式表现，例如其可以为手机、平板等。
图13为本实施例提供的一种电子设备的结构示意图。图13示出了适于用来实现本发明实施方式的示例性电子设备90的框图。图13显示的电子设备90仅仅是一个示例，不应对本发明实施例的功能和使用范围带来任何限制。
参见如图13，电子设备90的组件可以包括但不限于：至少一个处理器91、至少一个存储器92、连接不同***组件(包括存储器92和处理器91)的总线93。
总线93包括数据总线、地址总线和控制总线。
存储器92可以包括易失性存储器，例如随机存取存储器(RAM)921和/或高速缓存存储器922，还可以进一步包括只读存储器(ROM)923。
存储器92还可以包括具有一组(至少一个)程序模块924的程序工具925，这样的程序模块924包括但不限于：操作***、一个或者多个应用程序、其它程序模块以及程 序数据，这些示例中的每一个或某种组合中可能包括网络环境的实现。
处理器91通过运行存储在存储器92中的计算机程序，从而执行各种功能应用以及数据处理。
电子设备90也可以与一个或多个外部设备94(例如键盘、指向设备等)通信。这种通信可以通过输入/输出(I/O)接口95进行。并且，电子设备90还可以通过网络适配器96与一个或者多个网络(例如局域网(LAN)，广域网(WAN)和/或公共网络，例如因特网)通信。网络适配器96通过总线93与电子设备90的其它模块通信。应当明白，尽管图中未示出，可以结合电子设备90使用其它硬件和/或软件模块，包括但不限于：微代码、设备驱动器、冗余处理器、外部磁盘驱动阵列、RAID(磁盘阵列)***、磁带驱动器以及数据备份存储***等。
应当注意，尽管在上文详细描述中提及了电子设备的若干单元/模块或子单元/模块，但是这种划分仅仅是示例性的并非强制性的。实际上，根据本申请的实施方式，上文描述的两个或更多单元/模块的特征和功能可以在一个单元/模块中具体化。反之，上文描述的一个单元/模块的特征和功能可以进一步划分为由多个单元/模块来具体化。
实施例12
一种计算机可读存储介质，其上存储有计算机程序，所述程序被处理器执行时实现实施例1-5中任意一个实施例所述的语音打车方法的步骤。
其中，可读存储介质可以采用的更具体可以包括但不限于：便携式盘、硬盘、随机存取存储器、只读存储器、可擦拭可编程只读存储器、光存储器件、磁存储器件或上述的任意合适的组合。
在可能的实施方式中，本发明还可以实现为一种程序产品的形式，其包括程序代码，当所述程序产品在终端设备上运行时，所述程序代码用于使所述终端设备执行实现实施例1-5中任意一个实施例所述的语音打车方法的步骤。
其中，可以以一种或多种程序设计语言的任意组合来编写用于执行本发明的程序代码，所述程序代码可以完全地在用户设备上执行、部分地在用户设备上执行、作为一个独立的软件包执行、部分在用户设备上部分在远程设备上执行或完全在远程设备上执行。
虽然以上描述了本发明的具体实施方式，但是本领域的技术人员应当理解，这些仅是举例说明，在不背离本发明的原理和实质的前提下，可以对这些实施方式做出多种变更或修改。因此，本发明的保护范围由所附权利要求书限定。

Claims (16)

  1. 一种语音打车方法，其特征在于，所述语音打车方法包括：
    获取乘客的语音信息；
    从所述语音信息中提取目的地信息；
    判断与所述目的地信息对应的地址信息是否唯一，若否，则生成提示信息；所述提示信息用于提示再次获取所述乘客的语音补充信息；
    获取所述乘客的语音补充信息；
    从所述语音补充信息中提取目的地补充信息；
    根据所述目的地信息和所述目的地补充信息确定唯一对应的地址信息；
    根据所述唯一对应的地址信息生成打车订单。
  2. 如权利要求1所述的语音打车方法，其特征在于，所述判断与所述目的地信息对应的地址信息是否唯一，若否，则生成提示信息的步骤中，若判断为是，则确认所述目的地对应的地址信息为所述唯一对应的地址信息，然后执行所述根据所述唯一对应的地址信息生成打车订单的步骤。
  3. 如权利要求1-2中至少一项所述的语音打车方法，其特征在于，所述从所述语音信息中提取目的地信息的步骤之前，所述语音打车方法还包括：
    判断所述语音信息是否为打车信息，若是，则执行从所述语音信息中提取目的地信息。
  4. 如权利要求3所述的语音打车***，其特征在于，所述判断所述语音信息是否为打车信息，若是，则执行从所述语音信息中提取目的地信息的步骤具体包括：
    预设一语义库，所述语义库存储有打车标准文本；
    将所述语音信息转化为文本信息；
    将所述文本信息与所述打车标准文本进行比对，计算所述文本信息与所述打车标准文本的匹配度；
    判断所述匹配度是否大于预设的匹配阈值，若是，则确认所述语音信息为打车信息，然后执行从所述语音信息中提取目的地信息。
  5. 如权利要求1-4中至少一项所述的语音打车方法，其特征在于，所述从所述语音信息中提取目的地信息的步骤还包括：
    检测所述语音信息是否包括出发地信息，若是，则提取所述出发地信息，若否，则将所述乘客的当前位置信息作为所述出发地信息；
    所述根据所述唯一对应的地址信息生成打车订单的步骤中根据所述出发地信息和所述唯一对应的地址信息生成所述打车订单。
  6. 如权利要求1-5中至少一项所述的语音打车方法，其特征在于，所述判断与所述目的地信息对应的地址信息是否唯一，若否，则生成提示信息的步骤还包括：
    预存一地址库，所述地址库存储有地名信息与地理位置信息的映射；
    检测所述目的地信息是否包括所述地名信息，若是，从所述地址库中获取与所述地名信息对应的地理位置信息，并将所述与所述地名信息对应的地理位置信息作为所述唯一对应的地址信息，然后执行所述根据所述唯一对应的地址信息生成打车订单的步骤；若否，则判断与所述目的地信息对应的地址信息是否唯一。
  7. 如权利要求1-6中至少一项所述的语音打车方法，其特征在于，所述语音打车方法还包括：
    生成所述打车订单的确认请求；
    获取乘客根据所述确认请求发出的响应信息；
    识别所述响应信息，若所述响应信息为确认打车订单，则将打车订单发送至司机端；若所述响应信息为取消打车订单，则取消所述打车订单。
  8. 一种语音打车***，其特征在于，所述语音打车***包括语音信息获取模块、信息提取模块、目的地信息判断模块、提示模块、地址信息生成模块和打车订单生成模块；
    所述语音信息获取模块用于获取乘客的语音信息；
    所述信息提取模块用于从所述语音信息中提取目的地信息；
    所述目的地信息判断模块用于判断与所述目的地信息对应的地址信息是否唯一，若否，调用所述提示模块；
    所述提示模块用于生成提示信息；所述提示信息用于提示再次获取所述乘客的语音补充信息；
    所述语音信息获取模块还用于获取所述乘客的语音补充信息；
    所述信息提取模块还用于从所述语音补充信息中提取目的地补充信息；
    所述地址信息生成模块用于根据所述目的地信息和所述目的地补充信息确定唯一对应的地址信息；
    所述打车订单生成模块用于根据所述唯一对应的地址信息生成打车订单。
  9. 如权利要求8所述的语音打车***，其特征在于，所述目的地信息判断模块还用于在判断为是时，确认所述目的地对应的地址信息为所述唯一对应的地址信息，然后调用所述打车订单生成模块。
  10. 如权利要求8-9中至少一项所述的语音打车***，其特征在于，所述语音打车***还包括打车信息判断模块；
    所述打车信息判断模块用于判断所述语音信息是否为打车信息，若是，则调用所述信息提取模块提取所述目的地信息。
  11. 如权利要求10所述的语音打车***，其特征在于，所述打车信息判断模块包括语义库、文本信息转化模块、计算模块和匹配度判断模块；
    所述语义库存储有打车标准文本；
    所述文本信息转化模块用于将所述语音信息转化为文本信息；
    所述计算模块用于将所述文本信息与所述打车标准文本进行比对，计算所述文本信息与所述打车标准文本的匹配度；
    所述匹配度判断模块用于判断所述匹配度是否大于预设的匹配阈值，若是，则确认所述语音信息为打车信息，然后调用所述信息提取模块提取所述目的地信息。
  12. 如权利要求8-11中至少一项所述的语音打车***，其特征在于，所述语音打车***还包括信息检测模块和出发地信息确认模块；
    所述信息检测模块用于检测所述语音信息是否包括出发地信息，若是，则调用所述信息提取模块提取所述出发地信息，若否，则调用所述出发地信息确认模块；
    所述出发地信息确认模块用于将所述乘客的当前位置信息作为所述出发地信息；
    所述打车订单生成模块还用于根据所述出发地信息和所述唯一对应的地址信息生成所述打车订单。
  13. 如权利要求12所述的语音打车***，其特征在于，所述目的地信息判断模块包括地址库和地理位置信息获取模块；
    所述地址库存储有地名信息与地理位置信息的映射；
    所述信息检测模块还用于检测所述目的地信息是否包括所述地名信息，若是，则调用所述地理位置信息获取模块，若否，则调用所述目的地信息判断模块；
    所述地理位置信息获取模块用于从所述地址库中获取与所述地名信息对应的地理位置信息，并将所述与所述地名信息对应的地理位置信息作为所述唯一对应的地址信息，然后调用所述打车订单生成模块。
  14. 如权利要求8-13中至少一项所述的语音打车***，其特征在于，所述语音打车***还包括确认请求生成模块、响应信息获取模块、响应信息识别模块、信息发送模块和取消订单模块；
    所述确认请求生成模块用于生成所述打车订单的确认请求；
    所述响应信息获取模块用于获取乘客根据所述确认请求发出的响应信息；
    所述响应信息识别模块用于识别所述响应信息，若所述响应信息为确认打车订单，则调用所述信息发送模块，若所述响应信息为取消打车订单，则调用所述取消订单模块；
    所述信息发送模块用于将所述打车订单发送至司机端；
    所述取消订单模块用于取消所述打车订单。
  15. 一种电子设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，其特征在于，所述处理器执行所述计算机程序时实现权利要求1至7中任意一项所述的语音打车方法。
  16. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所述程序被处理器执行时实现权利要求1至7中任意一项所述的语音打车方法的步骤。
PCT/CN2018/109010 2017-12-20 2018-09-30 语音打车方法及*** WO2019119927A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711384195.4A CN109949800B (zh) 2017-12-20 2017-12-20 语音打车方法及***
CN201711384195.4 2017-12-20

Publications (1)

Publication Number Publication Date
WO2019119927A1 true WO2019119927A1 (zh) 2019-06-27

Family

ID=66994395

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/109010 WO2019119927A1 (zh) 2017-12-20 2018-09-30 语音打车方法及***

Country Status (2)

Country Link
CN (1) CN109949800B (zh)
WO (1) WO2019119927A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112163685A (zh) * 2020-09-11 2021-01-01 广州宸祺出行科技有限公司 一种基于语音ai的智能撮合行程的方法及***
CN113011198B (zh) * 2021-03-05 2022-07-22 北京嘀嘀无限科技发展有限公司 一种信息交互方法、装置及电子设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103680135A (zh) * 2013-12-31 2014-03-26 北京东方车云信息技术有限公司 一种提供打车服务的方法、装置及***
CN104216972A (zh) * 2014-08-28 2014-12-17 小米科技有限责任公司 一种发送打车业务请求的方法和装置
US20170116982A1 (en) * 2015-10-21 2017-04-27 Google Inc. Parameter collection and automatic dialog generation in dialog systems
CN106951468A (zh) * 2017-03-02 2017-07-14 腾讯科技（深圳）有限公司 对话生成方法及装置

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101617595B1 (ko) * 2015-09-10 2016-05-03 주식회사 에세텔 스마트 택시 어플리케이션 시스템 운용 방법

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103680135A (zh) * 2013-12-31 2014-03-26 北京东方车云信息技术有限公司 一种提供打车服务的方法、装置及***
CN104216972A (zh) * 2014-08-28 2014-12-17 小米科技有限责任公司 一种发送打车业务请求的方法和装置
US20170116982A1 (en) * 2015-10-21 2017-04-27 Google Inc. Parameter collection and automatic dialog generation in dialog systems
CN106951468A (zh) * 2017-03-02 2017-07-14 腾讯科技（深圳）有限公司 对话生成方法及装置

Also Published As

Publication number Publication date
CN109949800B (zh) 2021-08-10
CN109949800A (zh) 2019-06-28

Similar Documents

Publication Publication Date Title
CN111028827B (zh) 基于情绪识别的交互处理方法、装置、设备和存储介质
CN108346430B (zh) 对话***、具有对话***的车辆以及对话处理方法
CN110838288B (zh) 一种语音交互方法及其***和对话设备
CN110415679B (zh) 语音纠错方法、装置、设备和存储介质
CN107945792B (zh) 语音处理方法和装置
US11297016B1 (en) Intelligent conversational systems
US8005488B2 (en) Automatic service vehicle hailing and dispatch system and method
US7907705B1 (en) Speech to text for assisted form completion
CN111212190B (zh) 一种基于话术策略管理的对话管理方法、装置和***
US20190164540A1 (en) Voice recognition system and voice recognition method for analyzing command having multiple intents
CN111710336A (zh) 语音的意图识别方法、装置、计算机设备及存储介质
US8886532B2 (en) Leveraging interaction context to improve recognition confidence scores
CN110503948B (zh) 对话***和对话处理方法
CN111651571B (zh) 基于人机协同的会话实现方法、装置、设备及存储介质
CN103177721A (zh) 语音识别方法和***
CN110223134B (zh) 基于语音识别的产品推荐方法及相关设备
CN111739519A (zh) 基于语音识别的对话管理处理方法、装置、设备及介质
WO2019119927A1 (zh) 语音打车方法及***
CN110008325A (zh) 一种基于商用对话***的口语语言理解及改写方法
CN113011159A (zh) 人工座席监听方法、装置、电子设备及存储介质
US12020696B2 (en) Automatic synchronization for an offline virtual assistant
CN113436614B (zh) 语音识别方法、装置、设备、***及存储介质
CN105869631B (zh) 语音预测的方法和装置
US10819849B1 (en) Device, system and method for address validation
US20200327888A1 (en) Dialogue system, electronic apparatus and method for controlling the dialogue system

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: 18892047

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18892047

Country of ref document: EP

Kind code of ref document: A1