JPH01108837A - Voice dialing equipment - Google Patents

Abstract

PURPOSE:To attain correction by recognizing and discriminating a voice inputted from a microphone or a handset, generating a multiplex wave dial tone (DTMF) to confirm the result of recognition and outputting it as an acoustic signal, thereby discriminating the result of recognition soon when it is in error. CONSTITUTION:A voice signal inputted from a microphone or a handset 1 is recognized and discriminated by a recognition discrimination section 2 and the recognized and discriminated result is transferred to a dial signal sending section 3 and sent to a line as a dial signal in a proper timing. That is, in case of a general telephone circuit, the signal is sent as a dial pulse(DP) or a DTMF after hook-off. When the voice is recognized and discriminated as a dial number, a DTMF generator 4 is driven and an acoustic signal is outputted from a speaker or a receiver 5. Moreover, a ring tone consecutive time of the DTMF signal is set to nearly 100ms, then the difference of the color tone is discriminated. Thus, whether or not the recognized result is identical to the input voice is discriminated by the user by having only to ring a single tone.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a voice dialing device that sends out dialing signals based on the recognition results of input speech, and particularly to improvements in means for confirming recognized dialed digits.

[Conventional technology]

BACKGROUND ART Conventionally, as a means for a user to confirm dialing by voice input, there is a method of repeating the recognized dial digits using a synthesized voice. Another method is to make a single sound sound for a short time so that recognition can be confirmed.

[Problem that the invention seeks to solve]

However, the user cannot judge whether the recognition result is the same as the input voice by simply making a single sound.

Also, the recognition results are read back point by point using synthesized voice.
It has the disadvantage that it is bothersome for the user and even feels noisy at 2 o'clock.

[Means for solving problems]

The voice dialing device of the present invention includes a recognition determination unit that recognizes and determines voice input from a microphone or a transmitter;
a dial signal sending unit that sends out a dial signal based on the recognition result; and at least a DT for confirming the recognition result.
It is composed of a sounding means using a speaker or a receiver for generating an MF signal and outputting the DTMF signal as an acoustic signal.

〔Example〕

FIG. 1 is a block diagram of a first embodiment of the present invention. The audio signal input from the microphone or transmitter 1 is
The recognition determination unit 2 performs a recognition determination. The recognition and determination results are transferred to the dial signal sending section 3, and sent to the line as a dial signal at an appropriate timing. That is, in the case of a general telephone line, after going off-hook, the signal is sent as DP (dial dial tone) or DTMF (multifrequency dial tone). The dialing signal may be sent out either finally or all at once. Furthermore, it goes without saying that if it is a car telephone line, the signal will be sent in a format determined by that system.

In addition, if the voice is recognized as a dialed number, the DTMF generator 4 is driven, and the DTMF generator 4 is activated to generate 2, for example, 6 if it is "1".
A two-frequency synthesized sound of 97Hz and 1209Hz is generated,
'5'', a two-frequency synthesized sound of 770Hz and 1336Hz is generated.The DTMF signal is output as an acoustic signal from the speaker or receiver 5.Set the sound duration time of the DT■'' signal to about 100m8. The difference in tone is clearly distinguishable.

In addition, for commands other than dialed numbers, such as clearing inputted dialed numbers or sending out to the line as a dialed signal, a single tone is similarly sent to 100m5.
It may also be possible to make a sound.

Furthermore, by setting the frequency of the single sound to be a different frequency for each command, it becomes easier to determine the recognition result.

It is easy to provide the DTMF generator 4 with a single-tone generation function without having a special single-tone generator.

FIG. 2 is a schematic diagram of a second embodiment of the present invention. In addition to the DTMF generator 4, the sounding means in this example includes a synthesized speech generator 6 driven by the recognition determination section 2,
According to external settings, the switch 7 changes whether to output a DTMF signal (or single tone) to the speaker or receiver 5, or to output a synthesized voice that repeats a recognized word or a synthesized voice that responds to a command. Is possible.

It goes without saying that in a telephone having a function of transmitting a DTMF signal on one line, the DTMF generator 4 described above and the DTMF generator used in the dial signal transmitting section 3 may be used in common.

〔Effect of the invention〕

As explained above, the voice dialing device of the present invention converts the recognition result of the input voice into DTMF if it is a dial signal.
If it is a command word, the user is notified by sounding a single sound with a different frequency depending on each word for a short period of about 100 m, so if the recognition result is wrong, it can be immediately identified and corrected. Become. Furthermore, since the system does not have to be repeated in synthesized voice each time, the user does not feel burdened.

Until the user becomes accustomed to the voice dialing device, it is possible to make it easier to identify recognition errors by repeating the recognition results using synthesized speech, as disclosed in the second embodiment.

Below string mortar

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a first embodiment of the present invention. FIG. 2 is a block diagram of a second embodiment of the invention. 1: Transmitter or microphone, 2: Voice recognition determination unit, 3: Dial signal sending unit, 4: DTMF generator, 5: Receiver or speaker, 6: Synthetic voice generator,
7: Switcher. Figure 1 Figure 2

Claims (1)

[Scope of Claims] 1. The voice dialing device is provided with means for, based on the result of recognizing the input voice, to sound a DTMF signal corresponding to the dialed number for a short time if the recognition result is a dialed number. Characteristic voice dialing device. 2. In the voice dialing device according to claim 1, when the recognition result is a command word other than a dial number, the sounding means emits a single tone of a different frequency for a short time according to the command word. A voice dialing device characterized by having a function of making a sound. 3. In the voice dialing device according to claim 2, when the recognition result is a command word other than a dial number, the sounding means repeats the command word with a synthesized voice or responds to the command. A voice dialing device comprising means for answering with a synthesized voice. 4. In the voice dialing device according to claim 3, when the recognition result is a dialed number, the sounding means transmits a DTMF signal or dials with synthesized voice according to an external setting. A voice dialing device characterized by having a function of repeating numbers. 5. In the voice dialing device according to claim 4, when the recognition result is a command word, the sounding means generates a synthesized voice in addition to a single tone of a different frequency depending on an external setting. A voice dialing device characterized by having a function of repeating a command word or responding to a command with a synthesized voice.