JPH044646A - Automatic dialing system - Google Patents

Abstract

PURPOSE:To attain sure automatic dialing even under noise in which performance of a voice recognizing device is deteriorated by allowing a speech synthesizer to utter a called party identification name sequentially when a noise level inputted from a microphone exceeds a prescribed value and controlling dialing in response to the operating state of a switch in response thereto. CONSTITUTION:When the system starts its operation, voice is inputted from a microphone 1, a speech recognition device 3 recognizes the voice and outputs the result to a microcomputer 5, the noise level of the voice signal from the microphone 1 is always monitored by a noise level detector 4. When an output level of the noise level detector 4 exceeds a threshold level and the state is consecutive for a prescribed time, the speech recognition device 3 invalidates the system control and validates the control by a manual switch 2. The voice synthesizer 6 reads sequentially a keyword corresponding to the registered destinations. The user depresses a manual switch 2 when the destination to which a phone call is desired to be made is uttered.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an automatic dialing system for automatically dialing a telephone.

[Summary of the Invention] In the present invention, when the noise level of the input from the microphone is below a predetermined value, dialing is controlled according to the output from the speech recognition device, and the noise level of the input from the microphone is set to the predetermined value. When the number exceeds 1, dialing is controlled in accordance with the operating state of the manual switch in response to the destination identification names sequentially generated from the speech synthesizer.

[Prior Art] Conventional automatic dialing systems are known to use a voice recognition device, which uses a voice to pronounce a keyword that uniquely corresponds to a telephone number.

The system simplifies the process by specifying the destination phone number and performing automatic dialing.

In a system that uses such a voice recognition device as a main component to simplify dialing operations, when it is applied in a noisy environment, for example, inside a car, a problem arises in that the performance of the voice recognition device deteriorates.

One known method for dealing with such noise is to improve the performance of the speech recognition device itself against noise. In this case, there are methods such as separating speech and noise using the correlation of speech and removing the noise, and selecting parameters that are resistant to noise as parameters used for speech recognition.

[Problem to be Solved by the Invention] However, in an actual noisy environment, there is a limit to achieving sufficient performance in all situations. -As a result, automated dialing systems using voice recognition devices become inconvenient to use depending on the surrounding noise.

SUMMARY OF THE INVENTION An object of the present invention is to provide an automatic dialing system that can perform reliable automatic dialing without hindering operation even in a noisy environment where the performance of a speech recognition device is degraded.

[Means for solving problem i] In order to achieve such an object, the present invention includes a microcomputer, a microphone for collecting sound, and a system that recognizes the sound input from the microphone and inputs the sound into the microcomputer. A speech recognition device that outputs, a noise level detector that detects the noise level input from the microphone and outputs it to the microcomputer, a speech synthesizer that generates speech according to the output from the microcomputer, and is connected to the microcomputer. and a dial control section that controls the dialing of the telephone according to the output from the microcomputer. When the noise level from the noise level detector exceeds a predetermined value, the voice synthesizer sequentially generates the destination identification names, and The present invention is characterized in that the dial control unit is configured to execute dialing in accordance with the operating state of the switch in response to the above.

[Function] In the present invention, when the noise level of the input from the microphone exceeds a predetermined value, the voice synthesizer sequentially utters the destination identification name, and dialing is controlled according to the operating state of the switch in response. Therefore, even in a noisy environment where the performance of the voice recognition device is degraded, automatic dialing can be performed reliably without any hindrance to operation.

[Example] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram of an embodiment of an automatic dialing system according to the present invention, in which 1 is a microphone, 2 is a manual switch, 3 is a voice recognition device, 4 is a noise level detector, 5 is a microcomputer, and 6 is a voice synthesizer. 7 is a speaker, 8 is a dial control unit, and 9 is a telephone set. Note that the microcomputer 5 includes a CPU 51, a ROM 52. −
It consists of RAM53 etc.

FIG. 2 shows the configuration of an embodiment of the speech recognition device 3 shown in FIG. 1, including an amplifier 31, an automatic gain control circuit (AGC)
32, a bandpass filter (BPF) bank 33, a multiplexer (MPX) 34, an A/D converter 35, a pattern memory 36, and a microcomputer 37. The microcomputer 37 includes a CPU 38, a ROM 39, and a R
It consists of AM40 etc.

FIG. 3 shows the configuration of an embodiment of the noise level detector 4 shown in FIG. 1, in which 41 is an amplifier, 42 is an absolute value circuit,
43 is a smoothing filter, 44 is a reference voltage, and 45 is a comparator.

Figure 4 shows an example of the operation waveforms of each part in Figure 3.
(a) shows the output of the amplifier 41, (b) shows the output of the absolute value circuit 42, (c) shows the output of the smoothing filter 43, and (d) shows the output of the comparator 45.

FIG. 5 is a flow chart for explaining the operation of the automatic dialing system according to the present invention.

Hereinafter, the operation of the system according to the present invention will be explained based on these figures.

In FIG. 1, when the system starts operating, a voice is input from the microphone 1, the voice is recognized by the voice recognition device 3, and outputted to the microcomputer 5, while the noise level of the voice from the microphone 1 is changed to a noise level. It is constantly monitored by a level detector 4.

Specifically, in the speech recognition device 3, as shown in FIG. 2, after an input signal from the microphone 1 is amplified by an amplifier 31, an automatic gain control circuit 32, a bandpass filter bank 33 . A/D converter 35 via multiplexer 34
, pattern memory 36, and their A/D converter 3
5. Input the output of the pattern memory 36 to the microcomputer 37.

In the noise level detector 4, as shown in FIG. 3, after the input signal from the microphone 1 is amplified by an amplifier 41 (waveform in FIG. 4(a)), an absolute value circuit 42 and a smoothing filter 43 are used. (waveforms in Figure 4 (b) and (C) through),
Outputs the average energy level of the signal. The comparator 45 uses a preset reference voltage 44 and a smoothing filter 43.
The comparison result (waveform 9 in FIG. 4(d)) is output to the microcomputer 5.

The microcomputer 5 determines whether the output of the noise level detector 4 exceeds the threshold value (step 61 in FIG. When it is determined that the noise level is such that the noise level is reached, the system is controlled by the voice recognition device 3 (step 62 in FIG. 5). Conversely, when the output level of the noise level detector 4 exceeds the threshold,
That is, when it is determined that the performance of the voice recognition device W3 is at a level where it is degraded, the control of the system by the voice recognition device 3 is disabled, and the control by the manual switch 2 is enabled.

Here, as a method for determining the noise level, for example,
When the length of time during which the noise level exceeds the set threshold value exceeds a certain value (step 63 in FIG. 5), a method is used in which it is determined that the manual switch 2 should be used.

Below is a case where the speech recognition device 3 is enabled.

The operation of the system will be described when control by the manual switch 2 is enabled.

When the speech recognition device 3 is enabled, when the system starts, the microcomputer 5 uses the speech synthesizer 6,
Through the speaker 7, the user is notified that the mode is voice control mode and is prompted to input voice. When a keyword is input by voice, the voice recognition device 3 outputs the recognition result to the microcomputer 5. The microcomputer 5 controls the telephone 9 via the dial control section 8, and dials the telephone number corresponding to the keyword. These series of operations are equivalent to those of a conventional voice automatic dialing system.

When the control by the manual switch 2 is enabled, the output of the voice recognition device 3 is not used, and the destination for automatic dialing is specified by the user's operation of the manual switch 2 and the coordination with the voice synthesizer 6. This is done by

Specifically, when the output level of the noise level detector 4 exceeds the threshold and continues to do so for a certain period of time, the microcomputer 5 notifies the user through the voice synthesizer 6 and speaker 7 that the manual mode is active. At the same time (step 64 in FIG. 5), the speech synthesizer 6 sequentially reads out the keywords corresponding to the registered destination (step 64 in FIG. 5).
Figure step 65). The user specifies the destination by pressing the manual switch 2 when the destination to be called is read out (step 6 row 67 in FIG. 5).

To cancel the designation, press the manual switch 2 several times in succession (steps 67-68 in FIG. 5).

The microcomputer 5 specifies the destination for automatic dialing while controlling the above operations (step 69- in FIG. 5).
72). Once the destination is determined, dialing is performed using the dial control unit 8 using the corresponding telephone number (steps 73-74 in FIG. 5).

In the case of operation using the manual switch 2, it is necessary to use one's hands, but it is desirable to avoid an extremely poor usability compared to the case of performing voice recognition. For example, in applications such as car telephones,
If you want to be able to control automatic dialing while driving, this is possible with a voice recognition device. There is a need for an operation method that does not interfere with driving operations even when operating switches.

At this time, the most important thing is not to operate the switch visually. Therefore, if you operate as described above, there is only one switch to operate the switch, and there are only two ways to press it: press it once or press it multiple times, and you can operate the switch without visual confirmation. . Therefore, switch control that does not interfere with driving operations is possible.

[Effects of the Invention] As described above, according to the present invention, automatic dialing can be controlled reliably even in an environment with a noise level where the performance of the speech recognition device is degraded.

Furthermore, since a speech recognition device without noise resistance can be used, the cost of the device can be reduced.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an embodiment of an automatic dialing system according to the present invention, FIG. 2 is a block diagram showing an example of a specific configuration of the voice recognition device of FIG. 1, and FIG. A block diagram showing an example of a specific configuration of a level detector,
FIG. 4 is a waveform diagram of each part of FIG. 3, and FIG. 5 is a flowchart illustrating the operation of the automatic dialing system according to the present invention. 2・・・・・・・・・Manual switch, 3・・・・・・・・・
...Voice recognition device. 4...Noise level detector, 5...
・・・・・・Microcomputer, 6・・・・・・・・・
・Speech synthesizer, 8...Dial controller,
9...Telephone.

Claims (1)

[Claims] a microcomputer, a microphone that collects sound, a voice recognition device that recognizes the sound input from the microphone and outputs it to the microcomputer, and a voice recognition device that detects the noise level input from the microphone and outputs it to the microcomputer. a noise level detector that outputs a noise level detector, a voice synthesizer that generates sound according to the output from the microcomputer, a switch connected to the microcomputer, and a telephone dialing system that operates according to the output from the microcomputer. The microcomputer is equipped with a dial control unit that controls the
When the noise level from the noise level detector is below a predetermined value, the dial control section executes dialing according to the output of the voice recognition device, and when the noise level from the noise level detector reaches a predetermined value. When the number of destinations is exceeded, the speech synthesizer is configured to sequentially generate destination identification names, and the dial control unit is configured to execute dialing in accordance with the operating state of the switch in response to the generated identification names. An automatic dialing system characterized by: