JP6090027B2 - Voice command compatible information terminal with specific sound - Google Patents

Description

  The present invention relates to an information terminal, and more particularly to a technique for performing automatic control of an information terminal using a voice command.

  Conventionally, since the incoming call response processing has been performed manually, there is a problem of erroneous operation. As a technology for automatically performing proxy response processing for incoming calls, for example, Patent Document 1 discloses a technology in which a telephone is automatically put on hold when a hand is released from the telephone handset during a call.

JP-A-5-327838

However, the technique of Patent Document 1 may be put on hold by erroneously detecting that the hand has left the handset of the telephone even if it is not desired to put the call on hold. The present invention has been made in view of the above problems, and its object is to provide a automatically performing techniques proxy response processing of the telephone from the call between the conversational partner on the contents of the conversation around voice command is there.

In order to solve the above-mentioned problems, the present invention is an information terminal that performs a predetermined operation by a voice command, and that is input to the own terminal with specific voice registration means for registering a specific voice uttered by the user at the head of the voice command Specific voice detecting means for detecting the specific voice from voice, specific voice signal extracting means for extracting a series of specific voice signals following the detected specific voice, and specification for transmitting the extracted specific voice signal to a predetermined server Audio signal transmitting means, processing result receiving means for receiving the processing result data corresponding to the transmitted specific voice signal, command determining means for analyzing the received processing result data and determining a command related to the operation of the terminal. , characterized by having a a terminal control means for controlling the operation of the terminal in response to the determination command.

  According to the present invention, since the operation of the terminal is performed by a voice command by a voice uttered by the user, there is an effect of preventing an erroneous operation due to manual input or manual operation.

FIG. 1 is a schematic configuration diagram of a voice command compatible information terminal system according to an embodiment of the present invention. FIG. 2 is a schematic functional configuration diagram of the information terminal 1. Figure 3 is a diagram schematically illustrating an example of registration content of a particular voice storage unit 10 4. Figure 4 is a diagram schematically showing a registration content example of the processing content storage unit 1 11. FIG. 5 is a sequence diagram for explaining an operation example of the information terminal according to the present embodiment. FIG. 6 is a flowchart for explaining the specific voice registration operation of the information terminal 1. FIG. 7 is a flowchart for explaining the terminal control operation by the voice of the information terminal 1.

  Hereinafter, an embodiment of the present invention will be described.

FIG. 1 is a schematic configuration diagram of a voice command compatible information terminal system according to the present embodiment. As shown in the figure, the voice command compatible information terminal system according to the present embodiment includes an information terminal 1 connected to a network 2, a voice recognition server 3, and an information terminal 4 .

The information terminal 1 is connected to the voice recognition server 3 via the network 2. The information terminal 1 has a telephone function.
The information terminal 4 has a telephone function and is connected to the information terminal 1 via the network 2.

The voice recognition server 3 has a function of converting voice signals into text.

  FIG. 2 is a schematic functional configuration diagram of the information terminal 1.

  The information terminal 1 includes a communication control unit 101, a man-machine interface unit 102, a terminal control unit 103, a specific voice storage unit 104, a specific voice signal extraction unit 105, a specific voice search unit 106, a voice recognition server transmission unit 107, and a text reception unit. 108, an operation keyword extraction unit 109, a processing content determination unit 110, a processing content storage unit 111, and a call control unit 112.

  The communication control unit 101 performs a connection interface for connecting to the network 2 and overall communication control between the network 2 and the information terminal 1. The communication control unit 101 transmits the specific voice signal transmitted from the voice recognition server transmission unit 107 to the voice recognition server 3, and passes the text information received from the voice recognition server 3 to the text reception unit 108.

Here, the specific sound signal means a sound signal that follows a specific sound described later and continues until silence is detected.

The man-machine interface unit 102 is a user interface such as a receiver having a microphone / speaker for a user to talk, a dial key for receiving an input operation from the user, and an LCD / LED for displaying information to the user. When receiving the voice from the user, the man-machine interface unit 102 passes it to the terminal control unit 103. When the man-machine interface unit 102 receives a button input operation related to a call origination operation from the user, the man-machine interface unit 102 passes the button input information to the terminal control unit 103. In response to an instruction from the terminal control unit 103, the man-machine interface unit 102 performs display processing related to call connection such as ringing and voice communication output. Man-machine interface unit 102, if a particular voice registration start operation has been performed by the user, initiates a particular voice registration process, when receiving a specific audio from a user and passes the specific sound to the terminal control unit 103.

  Here, the specific voice means a voice command instructing to start acquisition of a specific voice signal.

When the terminal control unit 103 receives a voice from the user from the man-machine interface unit 102, the terminal control unit 103 passes the voice to the specific voice signal extraction unit 105 and the call control unit 112. When the terminal control unit 103 receives button input information from the man-machine interface unit 102, the terminal control unit 103 determines call connection instruction information from the button input information, and passes the determined call connection instruction information to the call control unit 112. The terminal control unit 103 receives a processing command related to each of the hold, transfer, and outgoing cases from the call control unit 112, and determines the content of the display processing related to the call connection such as ringing and the voice output instruction from the received processing command Then, the determined instruction content is passed to the man-machine interface unit 102. The terminal control unit 103 receives the specific voice received when the man-machine interface unit 102 starts the specific voice registration process, and passes the received specific voice to the specific voice storage unit 104.

The specific voice storage unit 104 registers the specific voice received from the terminal control unit 103 in the specific voice storage table 10 40 .

  FIG. 3 is a diagram schematically illustrating an example of registered contents in the specific voice storage unit 104.

As shown in the figure, the specific voice storage unit 104 stores a specific voice storage table 1040 . In the specific voice storage table 1040, a record 1043 of a specific voice signal to be sent to the voice recognition server is registered for each target specific voice. The record 1043 has a field 1041 that stores the registered specific voice, and a field 1042 that indicates acquisition range information of the specific voice signal to be sent to the voice recognition server.

  The specific voice signal extraction unit 105 requests the specific voice search unit 106 to perform a specific voice search for the voice sent from the terminal control unit 103. When the specific voice signal extraction unit 105 receives the search result from the specific voice search unit 106, if the search result is the specific voice, the voice following the specific voice is sent to the voice recognition server transmission unit 107 as the specific voice signal. send.

When a specific voice search is requested from the specific voice signal extraction unit 105, the specific voice search unit 106 refers to the specific voice storage table 1040 of the specific voice storage unit 104 and determines whether the voice is a registered specific voice. A search is performed, and the search result is passed to the specific audio signal extraction unit 105.

Speech recognition server transmission unit 107 stores in advance the address information of the speech recognition server 3, when the specific sound signal takes accept from a particular audio signal extraction unit 105, the addressed speech recognition server 3 via the communication control section 101 A specific audio signal is transmitted.

Text receiving unit 108 stores in advance the address information of the speech recognition server 3 and via the communication control unit 101 Ru preparative receives text information passed from the speech recognition server 3, the text information to the operation keyword extraction section 109 give.

  The operation keyword extraction unit 109 extracts an operation keyword from the text information received from the text reception unit 108 and passes the extracted operation keyword to the processing content determination unit 110.

When receiving the operation keyword from the operation keyword extraction unit 109, the process content determination unit 110 refers to the process content determination table 1110 of the process content storage unit 111 and searches for the process content corresponding to the operation keyword. Detected process content (hold, transfer, outbound) and instructs the processing of the call control unit 112.

The processing content storage unit 111 stores a processing content determination table 1110 . FIG. 4 is a diagram schematically illustrating an example of registered content in the processing content storage unit 111.

As illustrated, the processing content storage unit 111 stores a processing content determination table 1110 . In the processing content determination table 1110, a record 1113 of processing content information automatically performed by the information terminal 1 is registered for each target operation keyword. The record 1113 includes a field 1111 that stores an operation keyword to be extracted, and a field 1112 that stores the processing content automatically performed by the information terminal 1.

  The call control unit 112 performs general processing related to call connection at the time of general outgoing call, incoming call, and call termination, and overall voice processing during a call. The call control unit 112 performs processing according to an instruction from the processing content determination unit 110. The call control unit 112 passes to the terminal control unit 103 processing instructions relating to each of the hold, transfer, and outgoing cases.

  FIG. 5 is a sequence diagram for explaining an operation example of the information terminal according to the present embodiment.

Note that this sequence starts from a state in which a call between the information terminal 1 and the information terminal 4 is established and the call is in progress.

  The information terminal 1 monitors the user's voice during a call, and when a specific voice is detected from the monitored voice (S101), the information terminal 1 starts acquiring a specific voice signal (S102). When the information terminal 1 detects silence for 3 seconds or longer after the acquisition of the specific audio signal (S103), the information terminal 1 ends the acquisition of the specific audio signal (S104). Then, the information terminal 1 extracts the specific audio signal from the acquired audio signal after the acquisition of the specific audio signal is completed (S105). Then, after extracting the specific audio signal, the information terminal 1 transmits the specific audio signal information to the audio recognition server 3 (S106).

  When the voice recognition server 3 receives the specific voice signal information, the voice recognition server 3 converts the received specific voice signal into text (S107). Then, the voice recognition server 3 converts the specific voice signal into text, and transmits the converted text information to the information terminal 1 (S108).

Upon receiving the text information, the information terminal 1 searches for an operation keyword from the received text information (S109). When the operation terminal is detected from the text information , the information terminal 1 executes a corresponding process (in this embodiment, it is assumed that the operation keyword related to the hold process is detected). When the information terminal 1 detects the operation keyword related to the hold process, the information terminal 1 automatically performs the hold process (S110).

FIG. 6 is a flowchart for explaining the specific voice registration operation of the information terminal 1. This flow starts from a state where the specific voice registration start operation is executed by the user.

The man-machine interface unit 102 starts a specific voice registration process (S201), and determines whether or not a specific voice has been input (S202). If a particular sound is input (Yes in S202), it passes the voice input to a particular voice storage unit 104, to be entered (No at S202), and terminates the specific sound registration operation.

When the specific voice is input, the specific voice storage unit 104 receives the specific voice input from the man-machine interface unit 102 and registers the specific voice (S203) . After registering the specific voice , the specific voice registration operation is terminated.

  FIG. 7 is a flowchart for explaining the terminal control operation by the voice of the information terminal 1.

  This flow is started when the information terminal 1 and the information terminal 4 are in a call.

  The specific voice signal extraction unit 105 requests the specific voice search unit 106 to perform a specific voice search. The specific voice search unit 106 determines whether a specific voice is detected from the voice signal during a call (S301), and passes the determination result to the specific voice signal extraction unit 105. The specific sound signal extraction unit 105 starts acquisition of the specific sound signal if the detection of the specific sound can be confirmed from the determination result (YES in S301) (S302), and if the detection is not confirmed (NO in S301), the detection is performed. Continue processing.

  The specific audio signal extraction unit 105 determines whether or not silence for 3 seconds or more has been detected after starting the specific audio signal acquisition process (S302). If the detection can be confirmed (YES in S303), the specific audio signal is extracted. The signal acquisition process ends (S304), and if the detection cannot be confirmed (NO in S303), the detection process is continued.

  The specific audio signal extraction unit 105 passes the acquired specific audio signal to the voice recognition server transmission unit 107 after completing the specific audio signal acquisition process (S304).

  The voice recognition server transmission unit 107 transmits a specific voice signal following the detected specific voice to the voice recognition server 3 via the communication control unit 101 (S305).

Text receiving unit 108, after transmitting to the voice recognition server, intends rows of determining whether the received text information within 30 seconds (S306). (In YES S306) is if the received text information within 30 seconds, passing the text information to the operator keyword extraction unit 109, the operation keyword extracting section 109 extracts the operation keyword from the received text information, extracted The operation keyword is passed to the processing content determination unit 110. The processing content determination unit 110 to which the operation keyword has been passed determines whether or not the passed operation keyword is an operation keyword that is “pending” (S307). On the other hand, (NO in S306) If you did not receive text information within 30 seconds, the process returns to step 301.

Processing content determining unit 110, when the extracted operation keyword Ru der "pending" (YES in S307) notifies the processing contents to the call control unit 112 is "pending".

The call control unit 112 automatically performs hold processing from the processing content “hold” received from the processing content determination unit 110 (S308), and returns to step 301 after the processing is completed.

Processing content determination unit 110, when the extracted operation keyword was not a "hold" (NO in S307), a determination whether the operation keyword from the received text information is "transmission" (S309), Operation Keywords but if it is "outgoing" (YES in S309), the intends row a judgment of whether the corresponding phone number of the person is registered in the telephone directory search (S310).

Processing content determining section 110, when the extracted keyword is "outgoing", it is determined whether to read the name of the other party from the text information, the telephone number of the appropriate person in the phone book search has been registered, it is registered If so (YES in S310), the processing content “calling” and the telephone number of the other party of the call destination are passed to the call control unit 109. If not registered (NO in S310), the process ends and the process goes to step 301 . Return.

The call control unit 112, the processing content determination unit 110, when receiving the telephone number of the processing content of "transmission" the other party is automatically destination telephone number dialing line physician partner ( (S311), a call process for calling the other party of the input telephone number is performed (S312), and the process returns to the start after the process is completed.

Processing content determination unit 110, when the processing content of "Call" can not be determined from the extracted keyword (NO in S309), the determination keyword extracted from the received text information is whether a "transfer" If carried out (S313), the extracted keyword is "transfer" (YES in S313), the phone book makes a determination of whether the corresponding phone number of the person is registered in the search (S314), it can not be extracted keywords and if (NO in S313) ends the process returns to step 301.

Processing content determining section 110, Phonebook the corresponding telephone number of the person in the search, it is determined whether it is registered, if it can be confirmed to have been registered (YES in S314), the destination and the processing of the "transfer" of passing the phone number of the other party to the call controller 1 12, if no appropriate telephone number of the person is registered in the phonebook search (NO at S314), the process ends and returns to step 301.

The call control unit 112, the processing content determination unit 110, when receiving the treatment and number of transfer recipient of "transfer" is performed automatically dialing the destination party telephone number (S315 ), the transfer process to the other of the input telephone number (S316), after the processing returns to step 301.

  The embodiment of the present invention has been described above.

  In the present embodiment, when the information terminal 1 detects a specific voice from the call voice from the user, the information terminal 1 transmits a subsequent specific voice signal to the voice recognition server 3. The voice recognition server 3 converts the specific voice signal received from the information terminal 1 into text, and transmits the converted text information to the information terminal 1. After receiving the text information from the voice recognition server 3, the information terminal 1 extracts the operation keyword from the text information, and automatically performs “hold”, “call”, and “transfer” operations based on the operation keyword. Control.

  Therefore, according to the present embodiment, when the information terminal 1 automatically controls the telephone operation by voice, it is possible to prevent an erroneous operation due to a user's manual operation.

  Moreover, in this Embodiment, the information terminal 1 will start acquisition of a specific audio | voice signal, if a specific audio | voice is detected from a telephone call voice, and will complete | finish acquisition of a specific audio | voice signal if the silence of 3 seconds or more is detected.

  Therefore, according to the present embodiment, the information terminal 1 automatically extracts the specific voice signal from the voice, thereby realizing the automatic control of the telephone operation by the voice and preventing the erroneous operation due to the manual operation of the user. it can.

  Further, in the present embodiment, the information terminal 1 performs display processing related to call connection such as ringing and voice communication output according to the processing content determination result, and if the key input is detected, the processing content determination result. The operation of the own terminal is controlled according to the above.

  Therefore, according to this Embodiment, the information terminal 1 can control the own terminal according to a processing content determination result, and can prevent the erroneous operation by a user's manual operation.

1, 4: Information terminal, 2: Network, 3: Voice recognition server, 101: Communication control unit, 102: Man-machine interface unit, 103: Terminal control unit, 104: Specific voice storage unit, 105: Specific voice signal extraction unit , 106: specific voice search unit, 107: voice recognition server transmission unit, 108: text reception unit, 109: operation keyword extraction unit, 110: processing content determination unit, 111: processing content storage unit, 112: call control unit

Claims (3)

An information terminal that performs a predetermined operation by a voice command,
A specific voice registration means for registering a specific voice emitted by the user at the head of the voice command;
Specific voice detecting means for detecting the specific voice from voice input to the terminal;
Specific audio signal extracting means for extracting a series of specific audio signals following the detected specific audio;
Specific audio signal transmitting means for transmitting the extracted specific audio signal to a predetermined server;
Processing result receiving means for receiving processing result data corresponding to the transmitted specific audio signal;
Command determination means for analyzing the received processing result data and determining a command related to the operation of the terminal;
Terminal control means for controlling the operation of the terminal according to the determined command;
A voice command compatible information terminal with a specific sound. The information terminal according to claim 1,
With particular sound, wherein the specific set of speech signal following specific sound which the particular audio signal extracting means to extract is the audio signal to the detection of the silence over a certain time after a specific sound voices that the detected Voice command compatible information terminal. The information terminal according to claim 1 or 2,
The command determined by the command determination means is displayed on a display unit included in the terminal, monitors input of a specific key or an arbitrary key, and detects the key input, and then detects the key according to the determined command. A voice command compatible information terminal with a specific sound, characterized by controlling the operation of.

Images