JP2018033540A - Lingual position/lingual habit determination device, lingual position/lingual habit determination method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a lingual position/lingual habit determination device, a lingual position/lingual habit determination method and a program in which a lingual position/lingual habit can be determined in a non-invasion manner.SOLUTION: A lingual position/lingual habit determination device 1 includes: an audio input unit 2 for inputting audio data concerning the utterance of an utterer h; a measuring unit 3 for extracting the audio data serving as a determination object from the input audio data, and measuring a sound feature value (zero crossing rate and mel-frequency cepstrum) concerning a lingual position and a lingual habit in the extracted audio data; and an estimation unit 4 for estimating the lingual position or the lingual habit of the utterer h, based on the measured sound feature value (sound feature value vector).SELECTED DRAWING: Figure 1

Description

  The present invention relates to a tongue position / lingual tongue determining apparatus, a tongue position / lingual tongue determining method, and a program.

  The oral cavity is an organ used for vocalization, breathing, chewing and swallowing. Since maintaining the oral environment in a normal state is extremely important for physical health and the like, the oral environment has been conventionally measured (see, for example, Patent Document 1).

  Malocclusion is one of the factors that have a great impact on the oral environment and may impair physical health. About 25% of malocclusions are caused by oral habits (such as finger sucking and tongue wrinkles (tongue protrusions)). Therefore, in order to correct malocclusion, oral muscle function therapy (MTF) for improving oral habits by improving muscle balance of the tongue, lips and face has been performed.

International Publication No. 2001/078602

  The treatment of malocclusion varies depending on what causes the oral habit. For this reason, it has become important to analyze the function of the tongue and identify oral habits that cause malocclusion. In conventional medical practice, analysis of tongue function is performed using X-rays, palatogram pressure sensors, etc. in order to identify oral habits, so there is a risk of exposure during examination, and it is necessary to install instruments in the oral cavity There is. Therefore, a non-invasive method for identifying and determining the tongue position and tongue fold is desired.

  On the other hand, acoustic recognition technology using acoustic models is rapidly developing. However, an acoustic recognition technology that can determine the tongue position and tongue tongue has not yet been devised.

  An object of the present invention is to provide a tongue position / lingual tongue determination device, a tongue position / lingual tongue determination method, and a program capable of determining the tongue position / tongue in a non-invasive manner.

In order to achieve the above object, a tongue position / lingual tongue determining apparatus according to the present invention includes:
Based on voice data related to the utterance of the speaker, a measurement unit that measures acoustic features related to the tongue position and tongue tongue of the speaker;
An estimation unit for estimating the tongue position or tongue tongue of the speaker based on the measured acoustic features;
Is provided.

In this case, the measurement unit measures, as the acoustic feature quantity, the number of zero crossings, which is the number that the waveform of the input voice data intersects with a zero level or a constant section near the zero level,
The estimating unit estimates the tongue position or tongue tongue of the speaker based on the measured number of zero crossings;
It is good as well.

Further, the measurement unit measures a mel frequency cepstrum coefficient of the input voice data as the acoustic feature amount,
The estimation unit estimates the tongue position or tongue tongue of the speaker based on the measured Mel frequency cepstrum coefficient;
It is good as well.

The tongue position or tongue fold is stored in association with information on the acoustic feature for reference,
The estimation unit estimates a tongue position or tongue tongue associated with the reference acoustic feature quantity closest to the measured acoustic feature quantity as the tongue position or tongue tongue of the speaker;
It is good as well.

Moreover, the acoustic feature for reference is the acoustic feature vector for reference that includes the number of zero crossings and the mel frequency cepstrum coefficient obtained from the speech data related to the utterances of a plurality of speakers having the same tongue position or tongue tongue. Memorize as quantity information,
The estimation unit compares the acoustic feature quantity vector having the zero-crossing number and the mel frequency cepstrum coefficient obtained from the speech data related to the subject's utterance with the reference acoustic feature quantity vector. Estimating tongue position or tongue fold,
It is good as well.

The measurement unit extracts the sound data of the consonant section as sound data for determination.
It is good as well.

The measurement unit extracts voice data of a section in which the number of zero crossings of the voice data is equal to or greater than a threshold value as voice data of a consonant section;
It is good as well.

Tongue position / tongue determination method according to the second aspect of the present invention,
A measurement step of measuring an acoustic feature amount related to the tongue position and tongue tongue of the speaker based on voice data related to the speaker's speech;
An estimation step for estimating the tongue position or tongue tongue of the speaker based on the measured acoustic features;
including.

The program according to the third aspect of the present invention is:
Computer
A measurement unit that measures acoustic features related to the tongue position and tongue tongue of the speaker based on voice data related to the speaker's speech,
An estimation unit that estimates the tongue position or tongue tongue of the speaker based on the measured acoustic feature amount;
To function as.

  According to the present invention, since the tongue position / tongue of the speaker is estimated based on the acoustic feature amount obtained from the voice data related to the utterance of the speaker, the tongue position / tongue can be determined non-invasively. it can.

It is a block diagram which shows the function structure of the tongue position / lingual tongue determination apparatus which concerns on embodiment of this invention. It is a figure which shows an example of a tongue position and a tongue fold. 3 (A) and 3 (B) are diagrams showing the distribution of the number of zero crossings for each tongue position / lingual tongue. FIG. 4A to FIG. 4E are diagrams showing audio data of a consonant section in audio data. FIG. 5A and FIG. 5B are diagrams showing the distribution of the Mel frequency cepstrum fourth coefficient for each tongue position and tongue tongue. 6A to 6E are graphs showing the measurement results of the Mel frequency cepstrum coefficient for each tongue position / lingual tongue. It is a figure which shows the vector space by which the acoustic feature-value vector was 2-dimensionally converted. It is a block diagram which shows the hardware constitutions of the tongue position / lingual tongue determination apparatus of FIG. It is a flowchart of the determination process in a tongue position / tongue fold determination apparatus. It is a flowchart of a measurement process.

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

  As shown in FIG. 1, for example, a mobile phone, a smartphone, a recorder, or a personal computer that can input the voice of the speaker h can be used as the tongue / lingual tongue determining apparatus 1 according to the present embodiment.

  Tongue position and tongue tongue to be judged include “Origin”, “Lower tongue”, “Mandibular”, “Protruding tongue”, “Lower tongue” + Mandibular + Lower tongue ”. As shown in FIG. 2, “Origin” means that the tip of the tongue 22 is located immediately behind the upper front teeth 20, and the wide portion of the tongue 22 is lightly attached to the palate (the ceiling portion) of the upper jaw. This is the state of the tongue 22 shown by the solid line in FIG. “Lower tongue” is a state in which the position of the tip of the tongue 22 is low and the back side of the lower front tooth 21 is touched, that is, the state of the tongue 22 shown by a dotted line in FIG. “Mandibular” is a condition in which the teeth on the lower jaw (including the front teeth 21) protrude forward from the whole teeth (including the front teeth 20) on the upper jaw when meshed (see FIG. 2). The reverse state). “Protruding tongue” is a state in which the tongue 22 protrudes between the front teeth 20 and 21. "Lower tongue + lower tongue" (Mandibular + Lower tongue) is a condition in which lower tongue and lower jaw protrusion are combined.

  The tongue position / lingual tongue determination device 1 determines that the tongue position / tongue of the utterer h is “Origin”, “Lower tongue” based on the input voice data related to the utterance of the speaker h. ) ”,“ Mandibular ”,“ Protruding tongue ”, and“ Lower tongue + Mandibular + Lower tongue ”.

  As shown in FIG. 1, the tongue position / lingual tongue determination apparatus 1 according to the present embodiment is based on a voice input unit 2 that inputs voice data related to the utterance of the speaker h, and the input voice data. A measuring unit 3 that measures an acoustic feature amount related to the tongue position and tongue tongue of the speaker h, and an estimation unit 4 that estimates the tongue position or tongue tongue of the speaker h based on the measured acoustic feature amount; Is provided. The tongue position / lingual tongue determining apparatus 1 includes a storage unit 5 that is a storage device that stores various data.

  The voice input unit 2 is a microphone and inputs voice data related to the utterance of the speaker h. The input voice data is stored in the storage unit 5 as the voice data 10. The speaker h utters a certain fixed word, for example, the word “delicious”. The voice input unit 2 inputs the voice generated by this utterance as voice data. As words spoken by the speaker h, words including “S”, which is a consonant in which formants exist, are used. Here, the formant is a cluster of a plurality of peaks moving in time included in the spectrum of the voice of the person who is speaking. The inventor has found that the above-mentioned tongue position / tongue wall and formant have a high correlation.

  The measuring unit 3 measures an acoustic feature amount related to the tongue position and tongue tongue in the extracted voice data. Such acoustic features include the number of zero crossings of the voice level in the extracted voice data. That is, the measuring unit 3 measures the number of zero crossings, which is the number of times that the waveform in the extracted voice data intersects with a zero level or a constant section near the zero level.

  It is known that the number of zero crossings decreases in a vowel section and increases in a consonant section when, for example, vowels and consonants are uttered. Further, it is known that the number of zero crossings varies depending on the above-mentioned tongue position / tongue of the speaker even when a plurality of persons who have made the same sound are emitted. As shown in FIGS. 3 (A) and 3 (B), “Lower tongue”, “Mandibular”, “Lower tongue + Mandibular + Lower tongue”, “ With regard to “Origin” and “Protruding tongue”, the distribution of the number of zero crossings for each tongue position and tongue fold in the consonant was confirmed. It is very different.

  In FIG. 3A, the horizontal line in the middle of the strip is the average value for each tongue position and tongue tongue, the upper end of the strip is the value of the upper quarter, and the lower end of the strip is the lower order. The data value is 1/4. Further, the upper end of the vertical line is the highest value in each tongue position / tongue, and the lower end of the vertical line is the lowest value in each tongue position / tongue. In FIG. 3B, the horizontal axis represents the number of zero crossings, and the vertical axis represents the density (appearance probability) associated with each tongue position and tongue tongue. These data are all obtained from the speech data of the consonant section.

  In addition, the measurement unit 3 extracts audio data of a section to be determined from the audio data 10 stored in the storage unit 5. For example, the measuring unit 3 extracts the sound data of the consonant section as the sound data for determination. For example, the number of zero crossings described above can be used to determine the consonant section.

  A method for extracting speech data of a consonant section from the number of zero crossings is as follows. When audio data (waveform data) as shown in FIG. 4A is obtained, the spectrum of the waveform data is as shown in FIG. 4B. The measuring unit 3 detects a zero cross point (Z cross (Only trigger)) at which the waveform intersects the zero level from the waveform of the audio data shown in FIG. 4A (FIG. 4C). Here, the number of zero cross points (Z cross (Each frame)) per frame (unit time) is as shown in FIG. The measuring unit 3 uses the audio data (audio data in the section indicated by Z cross (Detected result)) of the portion where the number of zero cross points per frame is equal to or greater than the threshold (FIG. 4E) as audio for determination. Extract as data.

  Furthermore, the measurement unit 3 measures the mel frequency cepstrum coefficient (MFCC) of the audio data as an acoustic feature amount. Specifically, the measuring unit 3 measures the mel frequency cepstrum coefficient of the audio data for determination extracted as the audio data of the consonant section. The mel frequency cepstrum coefficient is an acoustic feature amount that represents vocal tract characteristics as in the cepstrum. Here, the cepstrum is the result of frequency conversion (for example, Fourier transform) by regarding the sound spectrum as a signal. Mel indicates that the coefficient is calculated in consideration of the characteristics of human speech perception.

The measuring unit 3 emphasizes the high frequency component of the waveform of the audio data for determination using the pre-emphasis filter. The pre-emphasis filter is used to clearly show vocal tract characteristics by enhancing high frequency components. For example, the following formula can be adopted as the filter calculation formula.
y (n) = x (n) -px (n-1)
Here, n is a natural number and a sampling number. Further, x (n) is the sound waveform data for determination, and x (n−1) is the value of the previous sound data. p is a pre-emphasis coefficient, and 0.97 is often used, but the value to be set is arbitrary. Y (n) is the output of the filter.

  Further, the measuring unit 3 performs fast Fourier transform (FFT) on the audio data in which the high frequency component is emphasized after applying the window function (Humming window) to obtain the amplitude spectrum of the audio data.

  Subsequently, the measurement unit 3 compresses the amplitude spectrum by applying a mel filter bank. A mel filter bank is, for example, a plurality of triangular bandpass filters arranged side by side, and is a filter bank that is equally spaced on the mel scale. The mel scale is a frequency axis reflecting human speech perception, and its unit is mel. That is, the band-pass filters of the mel filter bank are narrower as the frequency is lower, and wider as the frequency is higher. The number of bandpass filters is called the number of channels.

  Further, the measurement unit 3 regards the compressed numerical sequence as a signal and performs discrete cosine transform to obtain a cepstrum. The low-order component of the obtained cepstrum is the mel frequency cepstrum coefficient (MFCC), and the measurement unit 3 extracts the MFCC. The MFCC is expressed as MFCC (1) to MFCC (20) in order from the lowest order.

  For example, as shown in FIGS. 5 (A) and 5 (B), “Lower tongue”, “Mandibular”, “Lower tongue + Mandibular + Lower tongue” , “Origin”, “Protruding tongue”, the distribution of MFCC (4) (4th coefficient) for each tongue position and tongue tongue in the consonant, the tongue position and tongue tongue In addition, the distribution of MFCC (4) is greatly different. 5A and 5B are the same as those in FIGS. 3A and 3B.

  6 (A) to 6 (E), “Lower tongue”, “Mandibular”, “Mandibular + Lower tongue”, “Healthy ( The measurement results of MFCC (2) to MFCC (13) in “Origin” and “Protruding tongue” are shown. As shown in FIG. 6 (A) to FIG. 6 (E), the change patterns (profiles) of MFCC (2) to MFCC (13) are in good agreement in each tongue position and tongue tongue, and the profile is It is different between tongue position and tongue tongue.

  The estimation unit 4 estimates the tongue position or tongue tongue of the speaker h based on the measured number of zero crossings and the mel frequency cepstrum coefficient (MFCC). Specifically, the storage unit 5 uses, as reference data 12, information related to acoustic feature amounts obtained from speech data related to utterances of a plurality of speakers (different from the speaker h) having the same tongue position or tongue tongue. I remember it. The estimation unit 4 estimates the tongue position or tongue tongue associated with the reference data 12 closest to the information related to the measured acoustic feature amount of the speaker h as the tongue position or tongue tongue of the speaker h.

  More specifically, the reference data 12 stored in the storage unit 5 is the number of zero crossings obtained from speech data related to the utterances of a plurality of speakers (excluding the speaker h) having the same tongue position or tongue tongue. And a reference acoustic feature vector having mel frequency cepstrum coefficients (MFCC) as elements. The estimation unit 4 generates an acoustic feature vector having each of the zero crossing number and the mel frequency cepstrum coefficient (MFCC) obtained from the speech data related to the speech of the speaker h as reference data (reference acoustic feature value). Compared with (vector) 12, the tongue position or tongue tongue of speaker h is estimated.

  If the number of zero crossings and the MFCC (1) to MFCC (8) collected for each tongue position and tongue fold are used as reference elements, the reference acoustic feature vector (reference data 12) is converted into a two-dimensional plane and illustrated. As shown in Fig. 7, "Lower tongue", "Mandibular", "Lower tongue + Mandibular + Lower tongue", "Origin", "Tongue" The region where the vector exists for the “protruding tongue” is clearly distinguished. The estimation unit 4 measures the number of zero crossings and the MFCC (1) to MFCC (8) from the speech data related to the utterance of the speaker h, and to which region the acoustic feature quantity vector having these values as elements belongs. , The tongue position / tongue of the speaker h is estimated. For example, in the space shown in FIG. 7, when the acoustic feature quantity vector is in the region of “Lower tongue”, the tongue position / tongue of the speaker h is “Lower tongue”. It is estimated to be.

  As shown in FIG. 8, the tongue / lingual tongue determining apparatus 1 includes a control unit 31, a main storage unit 32, an external storage unit 33, an operation unit 34, a display unit 35, and an input unit 36 as hardware configurations. . The main storage unit 32, the external storage unit 33, the operation unit 34, the display unit 35, and the input unit 36 are all connected to the control unit 31 via the internal bus 30.

  The control unit 31 includes a CPU (Central Processing Unit) and the like. When this CPU executes the program 39 stored in the external storage unit 33, each component of the tongue position / lingual tongue determination apparatus 1 shown in FIG. 1 is realized.

  The main storage unit 32 is composed of a RAM (Random-Access Memory) or the like. The main storage unit 32 is loaded with a program 39 stored in the external storage unit 33. In addition, the main storage unit 32 is used as a work area (temporary data storage area) of the control unit 31.

  The external storage unit 33 includes a nonvolatile memory such as a flash memory, a hard disk, a DVD-RAM (Digital Versatile Disc Random-Access Memory), and a DVD-RW (Digital Versatile Disc ReWritable). In the external storage unit 33, a program 39 to be executed by the control unit 31 is stored in advance. Further, the external storage unit 33 supplies data used when executing the program 39 to the control unit 31 in accordance with an instruction from the control unit 31, and stores the data supplied from the control unit 31.

  The measurement unit 3 and the estimation unit 4 described above correspond to the control unit 31, and the storage unit 5 corresponds to the main storage unit 32 and the external storage unit 33.

  The operation unit 34 includes a pointing device such as a keyboard and a mouse, and an interface device that connects the keyboard and the pointing device to the internal bus 30. Information regarding the content operated by the operator is input to the control unit 31 via the operation unit 34. The operation of the voice input unit 2, the measurement unit 3, and the estimation unit 4 is started by an operation input from the operation unit 34.

  The display unit 35 is composed of a CRT (Cathode Ray Tube), an LCD (Liquid Crystal Display), or the like. When the operator inputs operation information, an operation screen is displayed. The display unit 35 displays, for example, the result of tongue position determination.

  The input unit 36 is composed of a microphone. The input unit 36 inputs ambient sound and outputs it to the internal bus 30 as sound data. The voice input unit 2 includes a control unit 31 and an input unit 36.

  In addition, you may have a communication interface which can communicate via a communication network. Audio data received via such a communication interface can also be determined.

  The various components of the tongue / lingual tongue determining apparatus 1 shown in FIG. 1 include a control unit 31, a main storage unit 32, an external storage unit 33, an operation unit 34, a display unit 35, and an input unit. The function is exhibited by being executed using 36 as a hardware resource.

  Next, the operation of the tongue position / lingual tongue determining apparatus 1 according to the present embodiment will be described. FIG. 9 shows a flowchart showing determination processing executed by the tongue position / lingual tongue determination apparatus 1.

  As shown in FIG. 9, the voice input unit 2 performs a voice input process of inputting voice data related to the utterance of the speaker h (step S1). The voice input unit 2 stores the input voice data as the voice data 10 in the storage unit 5.

  Subsequently, the measurement unit 3 extracts the audio data to be determined from the audio data related to the utterance of the speaker h, and measures the acoustic feature quantity related to the tongue position or tongue tongue in the extracted audio data. A process is performed (step S2).

  In step S2, as shown in FIG. 8, first, the measurement unit 3 reads the audio data 10 and performs zero-crossing measurement for measuring the number of zero-crossings that intersect the zero level for the waveform of the audio data 10. (Step S10). And the measurement part 3 performs the consonant area extraction which extracts the audio | voice data of the area where the number of zero crossings is more than a threshold value as audio | voice data of a consonant area (step S11). In this consonant segment extraction, the measurement unit 3 performs acoustic feature amount calculation for calculating the number of zero crossings and the values of MFCC (1) to MFCC (8) in the extracted voice data (step S12). In this acoustic feature quantity calculation step, the measurement unit 3 stores the calculated acoustic feature quantity in the storage unit 5 as acoustic feature quantity data 11.

  Returning to FIG. 9, the estimation unit 4 further performs an estimation step of estimating the tongue position or tongue tongue of the speaker (subject) h based on the acoustic feature data 11 stored in the storage unit 5 (step S <b> 3). ). Basically, the estimation unit 4 between the acoustic feature quantity data 11 (the acoustic feature quantity vector of the speaker h) and the reference acoustic feature vector for each tongue position / tongue as reference data 12 Each distance is calculated, and the tongue position / tongue corresponding to the acoustic feature vector having the shortest distance is output as a determination result. This determination result is displayed on a screen, for example, and can be presented to the speaker h or a doctor.

  In addition, as the reference data 12 stored in the storage unit 5, the voice input unit 2 inputs voices of a plurality of subjects whose tongue position and tongue tongue are obvious before the above-described processing is executed. What stored the acoustic feature-value which the measurement part 3 measured with respect to data as the reference data 12 in the memory | storage part 5 should just be used. Further, the reference data 12 composed of an enormous amount taken on a national average may be stored in the storage unit 5.

  As described above in detail, according to the present embodiment, the tongue position / tongue of the speaker h is estimated based on the acoustic feature amount obtained from the speech data related to the speech of the speaker h. Tongue position and tongue fold can be determined during invasion.

  In the above embodiment, the tongue position / lingual tongue is determined using the acoustic feature quantity vector having the number of zero crossings and the MFCC (1) to MFCC (8) as elements, but the present invention is not limited to this. For example, MFCC (9) or higher may be included in the vector element for the determination.

  Further, the calculation method of the mel frequency cepstrum coefficient is not limited to the above. For example, the harmonic component may be emphasized by a harmonic filter (high pass filter) different from the pre-emphasis filter. Further, other window functions such as a rectangular window, a Gaussian window, and a Hann window may be used without using a Hamming window. Further, frequency conversion may be performed using fast Fourier transform without using discrete cosine transform.

  Further, in the above embodiment, the speaker h utters a certain fixed word, for example, the word “delicious”, but the present invention is not limited to this. The words uttered by the speaker h may be other words including “S” which is a consonant in which a formant exists.

  In the above embodiment, “Origin”, “Lower tongue”, “Mandibular”, and “Protruding tongue” are used as the tongue position and tongue tongue. Judged. However, the present invention is not limited to this. Other tongue positions and tongue folds may be determined. For example, it is possible to extract a tongue fold that sandwiches the tongue between the upper and lower front teeth 20, 21.

  Further, it may be determined only whether “Origin” or “Lower tongue”. That is, you may enable it to determine a part among tongue position and tongue tongue mentioned above.

  In the above embodiment, the tongue position / tongue of the speaker h is determined by using the number of zero crossings of the voice data and the mel frequency cepstrum coefficient as the acoustic feature quantity. However, the present invention is not limited to this. For example, it is also possible to determine the tongue position / tongue of the speaker h only from the number of zero crossings. The distribution of the number of zero crossings of the speaker h is measured, and the distribution is compared with the distribution for each tongue position / tongue (reference data 12), and the tongue position / tongue having the closest distribution curve is used as the determination result. You may do it. As described above, even if there is only one acoustic feature amount, it is possible to determine the tongue position / tongue of the speaker h by a statistical method. Any other acoustic feature can be used as long as it can determine the tongue position / tongue of the speaker h.

  In the present embodiment, the tongue position / lingual tongue determining apparatus 1 includes the voice input unit 2, but the present invention is not limited to this. That is, the voice input unit 2 may not be provided. For example, a tongue position / lingual tongue determination device that determines the tongue position of audio data sent from a remote place may be used.

  Moreover, in the said embodiment, although the tongue position / lingual tongue determination apparatus 1 was made into the mobile phone, the smart phone, the voice recorder, or the personal computer, for example, it is not restricted to this. The tongue position / lingual tongue determining apparatus 1 may be a dedicated apparatus.

  In addition, the hardware configuration and software configuration of the tongue position / lingual tongue determining apparatus 1 are merely examples, and can be arbitrarily changed and modified.

  A central part that performs processing of the tongue position / lingual tongue determining apparatus 1 including the control unit 31, the main storage unit 32, the external storage unit 33, the operation unit 34, the display unit 35, the input unit 36, the internal bus 30, and the like As described above, can be realized using a normal computer system without using a dedicated system. For example, a computer program for executing the above operation is stored in a computer-readable recording medium (flexible disk, CD-ROM, DVD-ROM, etc.) and distributed, and the computer program is installed in the computer. Thus, the tongue position / lingual tongue determining apparatus 1 that executes the above-described processing may be configured. Further, the tongue position / lingual tongue determining apparatus 1 may be configured by storing the computer program in a storage device included in a server device on a communication network such as the Internet and downloading the computer program by a normal computer system.

  When realizing the function of a computer by sharing an OS (operating system) and an application program, or by cooperation between the OS and an application program, only the application program portion may be stored in a recording medium or a storage device.

  It is also possible to superimpose a computer program on a carrier wave and distribute it via a communication network. For example, a computer program may be posted on a bulletin board (BBS, Bulletin Board System) on a communication network, and the computer program distributed via the network. The computer program may be started and executed in the same manner as other application programs under the control of the OS, so that the above-described processing may be executed.

  Various embodiments and modifications can be made to the present invention without departing from the broad spirit and scope of the present invention. The above-described embodiments are for explaining the present invention and do not limit the scope of the present invention. In other words, the scope of the present invention is shown not by the embodiments but by the claims. Various modifications within the scope of the claims and within the scope of the equivalent invention are considered to be within the scope of the present invention.

  The present invention is useful for estimating a speaker's tongue position and tongue tongue.

  DESCRIPTION OF SYMBOLS 1 Tongue position / lingual tongue determination apparatus, 2 Voice input part, 3 Measurement part, 4 Estimation part, 5 Storage part, 10 Voice data, 11 Acoustic feature-value data, 12 Reference data, 13 Reference data, 20, 21 Front tooth, 22 Tongue, 30 internal bus, 31 control unit, 32 main storage unit, 33 external storage unit, 34 operation unit, 35 display unit, 36 input unit, 39 program, h speaker

Claims (9)

Based on voice data related to the utterance of the speaker, a measurement unit that measures acoustic features related to the tongue position and tongue tongue of the speaker;
An estimation unit for estimating the tongue position or tongue tongue of the speaker based on the measured acoustic features;
Tongue position / tongue determination device. The measurement unit measures, as the acoustic feature amount, the number of zero crossings, which is the number at which the waveform of the input voice data intersects with a zero level or a constant section near the zero level,
The estimating unit estimates the tongue position or tongue tongue of the speaker based on the measured number of zero crossings;
The tongue position / tongue determining apparatus according to claim 1. The measurement unit measures a mel frequency cepstrum coefficient of input voice data as the acoustic feature amount,
The estimation unit includes
Based on the measured Mel frequency cepstrum coefficient, estimating the tongue position or tongue tongue of the speaker,
The tongue position / tongue determining apparatus according to claim 1 or 2. The tongue position or tongue fold is stored in association with information on the acoustic feature for reference,
The estimation unit estimates a tongue position or tongue tongue associated with the reference acoustic feature quantity closest to the measured acoustic feature quantity as the tongue position or tongue tongue of the speaker;
The tongue position / tongue determining apparatus according to any one of claims 1 to 3. A reference acoustic feature vector having a zero-crossing number and a mel frequency cepstrum coefficient obtained from speech data related to the utterances of a plurality of speakers having the same tongue position or tongue fold as a reference acoustic feature amount Remember as information,
The estimation unit compares the acoustic feature quantity vector having the zero-crossing number and the mel frequency cepstrum coefficient obtained from the speech data related to the subject's utterance with the reference acoustic feature quantity vector. Estimating tongue position or tongue fold,
The tongue position / tongue determining apparatus according to claim 4. The measurement unit extracts the sound data of the consonant section as sound data for determination.
The tongue position / tongue determining apparatus according to any one of claims 1 to 5. The measurement unit extracts voice data of a section in which the number of zero crossings of the voice data is equal to or greater than a threshold value as voice data of a consonant section;
The tongue position / tongue determining apparatus according to claim 6. A measurement step of measuring an acoustic feature amount related to the tongue position and tongue tongue of the speaker based on voice data related to the speaker's speech;
An estimation step for estimating the tongue position or tongue tongue of the speaker based on the measured acoustic features;
Tongue position / tongue determination method. Computer
A measurement unit that measures acoustic features related to the tongue position and tongue tongue of the speaker based on voice data related to the speaker's speech,
An estimation unit that estimates the tongue position or tongue tongue of the speaker based on the measured acoustic feature amount;
Program to function as.

Images