JPS61246796A - Voice response switch - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field J] The present invention relates to a voice response switch, and more particularly to a voice response switch that operates by recognizing human voice.

[Background Art J] As shown in FIG. 6, a conventional upstream voice response switch includes a filter circuit 11 that passes a human body signal in a frequency band corresponding to voice, and a level detection circuit that detects the output level of the filter circuit 11. 12 and level detection circuit 12
The control circuit 13 compares the output of the level detection circuit 12 with a preset reference value and outputs a control signal when the output of the level detection circuit 12 is equal to or higher than the reference value, and a switch element 14 that is closed by the control signal. A device is provided in which it is determined that the input signal to the filter circuit 11 is an audio signal when the input level to the control circuit 13 is equal to or higher than a reference value.

In this circuit configuration, it is determined whether or not it is a voice only by determining the level of a specific frequency band. If it is an input signal, the switch element 14 will be activated even if it is a noise that is not a voice, resulting in a problem of malfunction. Also, even if audio is input, it will be switched to switch element 1.
Since the switch element 14 is actuated regardless of whether the sound is emitted for the purpose of actuating the switch element 4, there is an inconvenience that the switch element 14 may be actuated when the switch element 14 is not desired to be actuated. .

For this reason, as shown in FIG. 7, one idea is to use a voice recognition device 15 to compare the control voice stored in the storage unit 16 with the input voice, and to open or close the switch element 3 when the two match. However, when targeting unspecified speakers, there is a problem that a long time is required for arithmetic processing for voice recognition, making it difficult to control the switch element 14 in real time. The problem with the current level of technology is that the recognition rate is generally low and malfunctions are likely to occur.

Another big problem is that increasing the GA rate requires an increase in the amount of information and calculations, which further delays processing time. On the other hand, when targeting a specific speaker, the user needs to register his/her own voice in advance, which makes the process of using the system cumbersome.

[Object of the Invention] The present invention has been made in view of the above points, and its main purpose is to extract formants, which are dominant frequency components that characterize vowels in speech. death,
The system determines whether or not to activate a switch element based on the movement of the speech vector in a vector space formed by multiple 7t speech vectors, which can operate in real time and has a high recognition rate (and can be used by any specific speaker). The purpose of the present invention is to provide a voice-responsive switch.

[Disclosure of the Invention I Figure 5 shows an example of a vowel spectrum,
The dominant frequency component that characterizes a vowel, that is, the frequency component at the peak of the spectrum, is called a formant. Vowels usually have multiple formants, starting with the lowest frequency (17th orman) F,, J27 orman) F2, 13th formant F 3, etc.
It is called. Among these 7 oremans, the contribution rate of the 17th orman) F, and #27 orman) F2 is the highest, and the 1st formant FI and @27 orman)
F2 can be used to determine vowels with a fairly high degree of accuracy.

Here, if the Japanese vowel /a//i//u//e/10/ is shown as a vector on the PIF2 diagram with the first formant F1 on the horizontal axis and the second formant F2 on the vertical axis, , each vowel is represented by the range shown by the broken line in FIG. Formants vary considerably depending on each individual, and the ranges representing each vowel overlap to a large extent, but in general, the five vowels uttered by the same person in the same environment
The formants of vowels are approximately pentagonal on the PIF2 diagram, and it is known that even if the environment changes or the speaker changes, the relative positional relationship of the five vowels, that is, the pentagonal shape, will be maintained and will move in parallel. It is being therefore,
The relative position when a vowel changes, that is, the change vector, remains approximately constant even if the environment or speaker changes. In other words, the vowel /a/ component is (800Hz 1800Hz),
The component of vowel 10/ (500Hz.

1000 Hz), the components of the change vector from /a/ to 10/ are (-300Hz, -800Hz), and the components of the change vector remain approximately constant even if the environment or speaker is different. Accordingly, the present invention discloses a voice response switch in which a switch element is activated when a plurality of vowels are input and a vowel change vector is detected. In addition, in the following explanation, the first formant F
1 and the second formant F2 are used to perform voice recognition, but the third formant F may also be used to further increase the recognition rate. In this case, by representing the formant of each vowel on the F + -F z F 3 space using the third formant F as the third wheel, it is possible to represent the formant of each vowel on the space without forming overlapping parts. It is possible.

(Embodiment) As shown in FIG. 1, an input signal is input to a formant extraction circuit 1, and a 17th Orman F1 and a second formant F2 are extracted. The output of the formant extraction circuit 1 is input to a control voice discrimination circuit 2, and when it matches a preset control voice, a control signal is output. The output of the control voice discrimination circuit 2 is input to the switch element 3, and when a control signal is input to the switch element 3, the switch element 3 is opened or closed.

FIG. 2 shows an example of the formant extraction circuit 1.

The formant extraction circuit 1 includes a large number of band filter groups 11, each having a band of 200 Hz and having different pass frequencies. ~lln, each bandpass filter 11, an analog/digital conversion circuit 12 that converts the output signal of ~lln into a digital signal, and each bandpass filter 11. . . . an arithmetic circuit 13 consisting of a microprocessor that detects formants from the output level values of .about.tin. The bandpass filters 111 to fin each have a frequency of O-200Hz, 2
00-400Hz, 400-600Hz, -122
00-2400Hz, . -fin
It is set to include all frequencies in the audio band. The arithmetic circuit 13 has the first formant FI and the 27th formant FI.
Olman) F2 and outputs a phoneme change signal for determining whether the input voice has changed. Note that formant detection is performed using hardware using a circuit configuration, but it may also be performed using a software method such as a linear prediction method.

FIG. 3 shows an example of the control speech discrimination circuit 2, which includes a first vector holding circuit 22 that stores formants when a phoneme change signal is input;
When the phoneme change signal is input, the first vector holding circuit 22
A change vector is calculated by subtracting the 7t formant stored in the second vector holding circuit 23 from the formant stored in the first pexyl holding circuit 22. a change vector calculation circuit 24, a storage unit 25 that stores the range of change vectors between each two vowels when three arbitrary vowels are arranged in a predetermined order, and an output value and storage of the change vector calculation circuit 24. The output value of the change vector calculation circuit 24 is compared with the setting value stored in the storage unit 25.
The comparison determination circuit 26 outputs a match signal when the match signal is within a set range stored in the 1, and the control signal generating circuit 27 outputs a control signal when the match signal is continuously input. The control speech discrimination circuit 2 determines whether the change vector of the input signal belongs to the set range stored in the storage section 25 each time a phoneme change signal is input to the control speech discrimination circuit 2. When it is determined that the change vector between each phoneme of the input signal is within the set range of the change vector of the control voice stored in the storage section 25, a match signal is output from the comparison determination circuit 26. Note that the parts of the control voice discriminating circuit 2 except for the storage section 25 are configured using a microprocessor 20.

(Operation) The operation will be explained below.0 For example, assume that a control voice set to output a control signal is composed of three vowels /a/10//e/ arranged in order, and the storage unit 25 stores / ,/
As the range of change vector from to 10/(300±α
, Hz, 800±QmHz), as the range of the change vector to 10/gara/e/(120±Q 3 Hz w1
200±α, Hz) is set.

Here, the sensitivity is adjusted by appropriately setting the values of a1 to a. Now, the vowel /a/10//e/ is the first
Assuming that voice, Iji2 voice, and 3rd voice are input consecutively, the formant extraction circuit 1 detects the vector components of each voice on the F + -F2 plane, and also detects changes in vowels. A phonological change signal is generated at each point in time. At the time when the first voice is input, the control voice discriminating circuit 2 first converts the slip F component of the first voice into the first voice.
When the zero-order second voice stored in the vector holding circuit 22 is input and a phoneme change signal is obtained, the vector component of the first voice stored in the first vector holding circuit 22 is input to the second vector holding circuit 23. At the same time, the vector component of the second voice is stored in the first vector holding circuit 22. At this time, the change vector calculation circuit 24 calculates the component of the change vector from the amount of change between the vector component stored in the second vector holding circuit 23 and the vector component stored in the first vector holding circuit 22. The storage unit 25 has /
As a component of the change vector from a/ to 10/ (300
±CI, Hz, 800±ff2Hz) is stored, so the comparison/judgment circuit 26 compares whether the output of the change vector calculation circuit 24 is within this setting range stored in the storage section 25, and calculates the change vector. Calculation circuit 24
If it is determined that the output value of is within the setting range of the storage unit 25, it is determined that the input signal has changed from 8/ to 10/. The vector component of the second voice stored in the #1 vector holding circuit 22 is input to the second vector holding circuit 23, and the vector component of the third voice is input to the #1 vector holding circuit 22.
The change vector calculation circuit 24 calculates the component of the change vector from the second voice stored in the second vector holding circuit 23 to the third voice stored in the first vector holding circuit 22. The storage unit 25 stores (120±α,) as a component of the change vector from 10/ to /e/.

1200±a4) is stored, the comparison/judgment circuit 26 compares this setting range with the output value of the change vector calculation circuit 24, and determines whether the output value of the change vector calculation circuit 24 is within the setting range of the storage section 25. If it is determined that there is, it is recognized that the input signal has changed from 10/ to /e/. As mentioned above, the change from /a/ to 10/ and 1
When a change from 0/ to /e/ is detected continuously, the comparison/judgment circuit 26 outputs a coincidence signal, and the control signal generation circuit 27 receives the coincidence signal and outputs a control signal. The control signal is input to the switch element 3 and the control signal is input to the switch element 3.
is opened and closed. When the input signal differs from the setting range set in the storage section 25, the switch element 3 maintains its previous state.

In the above embodiment, the switch element 3 is opened and closed when three consecutive vowels are detected, but the present invention is not limited to three vowels. In addition, in order to detect a vowel, a vector in a two-dimensional space was used with the first formant F1 and the second formant F2 as vector components. Vowels may be determined using vectors in a multidimensional space of more than 3 dimensions.Furthermore, in the above embodiment, the formant extraction circuit 1
Although an example has been shown in which microprocessors are used for each of the circuits 1 and 2, the microprocessors for both circuits 1 and 2 may be shared and integrated into one.

[Effects of the Invention] As described above, the present invention includes a formant extraction circuit that extracts at least a first formant and a second formant from input speech, and a formant extraction circuit that extracts at least a first formant and a second formant from an input speech, and a formant extraction circuit that extracts at least a first formant and a second formant from an input speech, and a It consists of a control speech discrimination circuit that outputs a control signal when the change in formant is within a predetermined setting range, and a switch element that is opened and closed by the control signal, so it can characterize vowels in speech. By extracting the formant, which is the dominant frequency component of the vowel, and determining whether to activate the switch element by moving the speech vector in the vector space formed by multiple formants, we can detect only the change in the formant of the vowel. It only needs to be detected, the amount of calculation is small, and speech recognition can be performed reliably, and it has the advantage of being able to operate in real time and having a high recognition rate. Furthermore, since speech is recognized based on changes in formants, it has the advantage of being operable for unspecified speakers.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a block diagram showing a formant extraction circuit used in the above, FIG. 3 is a block diagram showing a control speech discrimination circuit used in the same, and FIG. The figure is an operation explanatory diagram showing an example of PIF2 diagram,
FIG. 5 is an operation explanatory diagram showing an example of vowel frequency characteristics, FIG. 6 is a block diagram showing a conventional example, and FIG. 7 is a block diagram showing another conventional example. 1 is a formant extraction circuit, 2 is a control voice discrimination circuit, and 3 is a formant extraction circuit.
is a switch element.

Claims (1)

[Claims] (1) At least the first formant and the second formant from the input voice
a formant extraction circuit that extracts a formant;
A control voice discrimination circuit that outputs a control signal when the change in formant between each vowel of a control voice composed of continuous vowels of more than one vowel is within a predetermined setting range, and a switch element that is opened and closed by the control signal. A voice response switch comprising: