JPS6382155A - Telephone set - Google Patents

Abstract

PURPOSE:To accurately recognize the voice of user by using a voice storage section so as to store a response voice of the user in requesting a voice response to the user by means of a synthesized voice and giving a voice by the user in response to the user and outputting the voice in a voice storage section to a general terminal control section after all the synthesized voice is outputted. CONSTITUTION:In requesting a voice response by the user by means of the synthesized voice and inputting a voice by the user in response to the request, a switch 11 is set to as to input the input voice to the voice storage section 10. Then the voice storage section 10 stores the response voice of the user for a prescribed time and after all the synthesized voice is outputted, the voice in the voice storage section 10 is outputted to the general terminal control section 5. Thus, even when the user applies voice reply during the output of the synthesized voice, the voice of the user is inputted accurately to the voice recognition section 2 to attain the recognition processing.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a telephone device having a voice recognition section and a voice synthesis section for controlling multifunctional voices used in the telephone device.

BACKGROUND OF THE INVENTION In recent years, telephone devices capable of various functions such as three-party calling and missed optical transfer have been introduced into the market and are attracting attention. In order to enable multiple functions, early telephone devices had a one-to-one correspondence between dial numbers and functions, and the user operated the functions by inputting the function number by dialing. was. However, as functions have become more diverse, it has become difficult to memorize function numbers, and the operation of the die 3 pager input has become complicated and difficult to handle. Therefore, in order to solve the above problems, conventional telephone devices have a voice recognition section and a voice synthesis section in the telephone device, and notify all users when the function is effective using synthesized voice.
The user inputs the function name aloud according to the synthesized voice, and when the user's voice is recognized, the full function is activated. The above method makes it possible to easily handle multiple functions without having to memorize all function numbers.

Hereinafter, the conventional telephone device as described above will be explained with reference to all the drawings. FIG. 3 is a block diagram of a conventional telephone device. In FIG. 3, 1 is a speech synthesis section;
2 is a voice recognition unit, 3 is a special terminal control unit, 4 is a system control unit, 12 is a general terminal control unit, 13 is a hybrid coil, 14 is a voice input terminal, 8 is a voice output terminal, and 15 is a number input terminal. . The operation of the telephone device configured as described above will be explained.

When a user uses each function of the telephone device, the user first inputs a function start number from the number input terminal 15, and the input signal is input to the system control unit 4 via the general terminal control unit 12. be done.

Next, the system control section 4 requests a synthesized speech from the speech synthesis section 1 via the special terminal control section 3, and the speech synthesis section 1 outputs the synthesized speech for requesting the user for the function name. The synthesized speech may be, for example, "Please give me the function name." The synthesized voice is generated by the special terminal control unit 3. System control unit 4. The general terminal control unit 12 is outputted from the audio output terminal 8 via the hybrid coil 13. In response to the synthesized voice, the user inputs the desired function name voice from the voice input terminal 14, and the function name voice is input to the hybrid coil 13. General terminal control unit 1
2. System control unit 4. It is input to the speech recognition section 2 via the special terminal control section 3, and after recognition processing, the recognition result is output to the special terminal control section 3. For example, the feature name is "Call Forwarding"
Suppose it was.

Next, the special terminal control section 3 requests the speech synthesis section 1 for synthesized speech for confirming the recognition result, and the speech synthesis section 1 outputs the synthesized speech. The synthesized voice may be, for example, "Is this call forwarded?" The synthesized speech is output from the speech output terminal 8 in the same way as the synthesized speech for requesting the page name in function 6. In response to the synthesized speech, the user inputs speech for determining whether the recognition result is correct or incorrect from the speech input terminal 14. It is assumed that the correct/incorrect judgment voice is, for example, “yes” or “no”. The correct/incorrect judgment voice is input to the voice recognition unit 2 in the same way as the function name voice, and after recognition processing, the recognition result is output to the special terminal control unit 3.

In the case of a correct recognition result, a signal for operating a function is sent from the special terminal control unit 3 to the system control unit 4, and in the case of an incorrect recognition result, a signal is sent from the special terminal control unit 3 to the system control unit 1. , requests the synthesized voice for requesting the function name again.

Problems to be Solved by the Invention However, in the above configuration, since the voice line between the general terminal control unit 12 and the hybrid coil 13 is two wires, the synthesized voice and the user's response voice are communicated simultaneously. That is impossible. For this reason, if the user starts a voice response while the synthesized voice is being output, the voice response will not be output to the general terminal control unit 12 exactly 6/2, making voice recognition impossible. In the conventional operation described above, synthesized speech and voice response are usually not input at the same time, but when all of the above conventional devices are actually used, it is possible for the user to input voice response without waiting for the synthesized voice to finish. This has caused the problem that voice recognition becomes impossible each time it occurs.

In view of the above-mentioned problems, the present invention stores the user's response voice in the voice storage unit and synthesizes it when the user is requested to give a voice response using synthesized voice and the user inputs voice in response to the request. By outputting the voice in the voice storage unit to the general terminal control unit after all voices are output, even if the user makes a voice response while the synthesized voice is being output, all of the user's voices can be accurately recognized. The present invention provides a telephone device that can perform recognition processing by inputting information into a section.

Means for Solving the Problems In order to achieve this object, the telephone device of the present invention includes a voice recognition unit that recognizes the user's voice, and a voice recognition unit that requests voice input from the user and performs recognition by the voice recognition unit. Result 7: a speech synthesis unit that controls synthesized speech to notify all users of the page result; a special terminal control unit that controls the speech recognition unit and speech synthesis unit; and a speech output that outputs the synthesized speech and the user's voice. a terminal, an audio input terminal for inputting the user's voice,
A number input terminal for inputting a number, an audio storage unit that stores all of the user's voice, a general terminal control unit that controls the telephone terminal that incorporates each of the above terminals and the audio storage unit, and each of the above controls. and a system control section that controls the system.

Effect: With this configuration, when a voice response is requested from the user using synthesized voice and the user inputs voice in response to the request, the user's response voice is stored in the voice storage section, and all the synthesized voice is output as voice. After being output from the terminal, the audio in the audio storage section is output to the membrane terminal control section.

Embodiment Hereinafter, the first example of a telephone device according to an embodiment of the present invention will be described.
This will be explained with reference to the figures. In FIG. 1, 1 is a speech synthesis section, 2 is a speech recognition section, 3 is a special terminal control section, 4 is a system control section, 5 is a general terminal control section, 6 is a hybrid coil, 7 is an audio input terminal, and 8 is a An audio output terminal, 9 a number input terminal, 10 an audio storage section, 11 a switch, and the same parts as in the conventional example shown in FIG. 2 are given the same numbers.

The operation of the telephone device configured as described above will be explained below using the flowchart of an example of voice conversation shown in FIG.

When the user uses all the functions of the telephone device.

The user inputs the function start number from the number input terminal 9 (conversation example 2o), and the input signal is sent to the voice storage section 10.
The signal is input to the system control unit 4 via the hybrid coil 6 and the general terminal control unit 6, and the system control unit 4 requests the speech synthesis unit 1 via the special terminal control unit 3 for synthesized speech, and requests the function name from the user. The synthesized speech for this purpose is output from the speech synthesis section 1. The synthesized speech may be, for example, "Please give me the function name" (conversation example 21). The synthesized voice is produced by special terminal 9.
Page control section 3. System control unit 4. The signal is output from the audio output terminal 8 via the general terminal control unit 6 and the hybrid coil 6. Further, the voice storage unit 10 that receives the number signal sets the switch 11 to the voice storage unit 10 side (terminal A) in order to store the user's input voice from now on.

Next, the user answers the synthesized voice and inputs the desired function name voice from the voice input terminal 7. For example, assume that the function name is "call forwarding" (conversation example 22). Function name: Before the sound is output to the hybrid coil 6,
After being output to the audio storage unit 1o and stored in the audio storage unit 10 for a certain period of time, the hybrid coil 6, the general terminal control unit 5. System control unit 4. The signal is input to the speech recognition section 2 via the special terminal control section 3. Preferably, the storage time in the audio storage unit 1o is slightly shorter than the synthesized audio length.

For example, if the synthesized voice is "1 function name please", approximately 1
Seconds is appropriate. After the recognition process, the recognition result is output to the special terminal control section 3, and the special terminal control section 3 then sends it to the speech synthesis section 1.
Then, the synthesized speech for confirming the recognition result is requested from the speech synthesizer 1, and the synthesized speech is output from the speech synthesis section 1.

The synthesized voice may be, for example, "Is this a call forwarding?" (conversation example 23). The synthesized speech is output from the speech output terminal 8 in the same way as the synthesized speech for requesting the function name. The user answers the synthesized voice and inputs the voice to judge whether the recognition result is correct or incorrect at the voice input terminal 7.
Enter more information. It is assumed that the correct/incorrect judgment voice is, for example, “yes” and “no” (conversation example 24). The correct/incorrect judgment sound is stored in the sound storage unit 1o for a certain period of time, similar to the function name sound.
After being stored in , it is input to the speech recognition section 2 , and after recognition processing, the recognition result is output to the special terminal control section 3 .

In the case of a correct recognition result, a signal for operating the function is sent from the special terminal control unit 3 to the system control unit 4, and the operation of the function is started (conversation example 26). In case of incorrect recognition results,
The special terminal control section 3 requests the speech synthesis section 1 again for synthesized speech requesting the function name.

When the above post-processing functional operations are completed, the system control unit 4
Then, an operation end signal is outputted to the audio storage section 10 via the general terminal control section 5. Next, the audio storage section 11 page.

At step 10, the switch 11 is set to the hybrid coil 6 side (B terminal side). In other words, the switch 11 operates to input the input voice to the input voice 7, input voice coil 6 during a normal call, and operates to input the voice to the voice storage section 1o when using each function.

As described above, according to this embodiment, when a synthesized voice is used to request a voice response from the user and the user inputs voice in response to the request, a switch is provided so that the input voice can be input into the voice storage section 1o. 11, the voice storage unit 10 stores the user's response voice for a certain period of time, and after all the synthesized voices are output, the voice in the voice storage unit 1o is output to the general terminal control unit 5. Even if the user makes a voice response while the synthesized voice is being output, the user's voice can be accurately input to the voice recognition unit 2 and recognition processing can be performed.

Note that in this embodiment, the storage time in the audio storage unit 10 is fixed at a certain time, but the storage time may be set as the time until each synthesized voice is outputted. The operation at this time will be explained below using the flowchart of the example voice conversation shown in FIG.

When the user uses the multi-functions of the telephone device, the user should input a function start number from the number input terminal 9 (conversation example 20), and the input signal is stored in the voice storage section 10. It is input to the system control unit 4 via the general terminal control unit 6, the system control unit 4 requests the speech synthesis unit 1 to synthesize speech via the special terminal control unit 3, and the speech synthesis unit 1 requests the user for the function name. A synthesized voice for the purpose is output. The synthesized speech may be, for example, "Please give me the function name." (Example Conversation 21).

The synthesized voice is generated by the special terminal control unit 3. System control section 4, general terminal control section 5. The signal is output from the audio output terminal 8 via the hybrid coil 6'z. When all the synthesized voices have been output, a synthesized voice end signal is input from the special terminal control section 3 to the system control section 4j and the general terminal control section 5 to the voice storage section 1o. On the other hand, in the voice storage section 10 that receives the number signal from the number input terminal 9, in order to save the user's input voice from now on, the switch 11 is switched on the side of the entire voice storage section 10 (A terminal 13
page side).

Next, the user inputs the desired function name voice from the voice input terminal 7 in response to the synthesized voice. For example, in the feature name "
Suppose that the message was "Non-call transfer" (Conversation example 22). The function name voice is output to the voice storage unit 10 for -1 time before being output to the NoA hybrid coil 6. At this time, if the synthesized voice end signal has already been output to the voice storage unit 1o, the function name The function name voice is directly output to the general terminal control section 5 without storing the voice. However, if the end signal has not been output yet, the function name voice is temporarily stored in the voice storage unit 10, and when the end signal is output, the function name voice is transmitted to the hybrid coil 6, the general terminal control unit 5, the system control Part 4.
The signal is input to the speech recognition section 2 via the special terminal control section 3.

After the recognition process, the recognition result is output to the special terminal control unit 3,
The special terminal control section 3 requests the speech synthesis section 1 for synthesized speech for checking the recognition result, and the speech synthesis section 1 outputs the synthesized speech. The synthesized voice may be, for example, "Is this a call forwarding?" (conversation example 23). 14 Heh.

The synthesized speech is output from the speech output terminal 8 in the same way as the synthesized speech requesting the function name. In response to the synthesized speech, the user inputs speech for determining whether the recognition result is correct or incorrect from the speech input terminal 7. It is assumed that the correct/incorrect judgment voice is, for example, “yes” or “no” (conversation example 24). The correct/incorrect judgment voice is stored in the voice storage unit 1o under the same conditions as the function name voice, and then input to the voice recognition unit 2. After recognition processing, the recognition result is output to the special terminal control unit 3.

In the case of a correct recognition result, a signal for operating the function is sent from the special terminal control section 3 to the system control section 4, and the operation of the function is started (conversation example 25). In the case of an erroneous recognition result, the special terminal control section 3 operates to request the speech synthesis section 1 again for a synthesized speech requesting the function name.

In this embodiment, in addition to the effects of the embodiments described above, it is possible to eliminate waste of time when storing response voices and shorten processing time.

Effects of the Invention The present invention provides that when a synthesized voice is used to request a voice response from a user and the user inputs a voice in response to the request, the voice storage unit stores the user's response voice, and the voice response is stored in the synthesized voice. By outputting the voice in the voice storage unit to the general terminal control unit after all the voices have been output, even if the user makes a voice response while the synthesized voice is being output.

The objective is to realize an excellent telephone device that can accurately input the user's voice to the voice recognition unit and perform recognition processing.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a telephone device according to an embodiment of the present invention, FIG. 2 is a flowchart of an example of the same voice conversation, and FIG. 3 is a block diagram of a conventional telephone device. 1...Speech synthesis unit, 2...Speech recognition unit, 3...Special terminal control unit, 4...System control unit, 5...・General terminal control unit, 6...
...Hybrid coil, 7...Audio input terminal, 8...Audio output terminal, 9...Number input terminal, 1o...Audio storage section, 11・・・・・・
·switch. Name of agent: Patent attorney Toshio Nakao and 1 other person No. 1
Figure 7---t? Input cuff #I) 8-, ±j3″ 9--Ichiban, No. △ force ・I View story call Song bowl 2nd figure 匡 === Sword - Licho wearer Ax Seisei low StM8 Irybetsu door Bbi b+p Fig. 3 8-1 S 4 A shout --- I Input ・1 15- No. 5... ・ Speak 7 to the mackerel fishing machine

Claims (1)

[Claims] When controlling the multi-functions used in a telephone device by voice response, there is a voice recognition unit that recognizes the user's voice, and a voice recognition unit that requests voice input from the user and sends the recognition results from the voice recognition unit to the user. a speech synthesis unit that controls synthesized speech to inform the user; a special terminal control unit that controls the speech recognition unit and the speech synthesis unit; an audio output terminal that outputs the synthesized speech and the user's voice; A voice input terminal for inputting a number, a number input terminal for inputting a number, a voice storage section for storing the user's voice, and a two-wire type and a four-wire type for branching the voice output terminal and voice input terminal. a general terminal control unit that controls a telephone terminal incorporating a hybrid coil that performs mutual conversion, the number input terminal, voice input terminal, voice output terminal, voice storage unit, and hybrid coil, and each of the above control units. and a system control unit that controls the user, and when the user requests a voice response using synthesized voice and the user inputs voice in response to the request, stores the user's response voice in the voice storage unit, A telephone device characterized in that after all synthesized speech is output from the speech output terminal, the speech in the speech storage section is output to the general terminal control section.