JP6775218B2 - Swallowing information presentation device - Google Patents

Description

The present invention relates to a swallowing information presenting device.

For example, when the swallowing function deteriorates, aspiration is more likely to occur when eating than usual. Therefore, for example, in the field of medical care or long-term care, a nurse or a caregiver may be required to confirm that swallowing is properly performed during the meal of the subject. Therefore, a means for easily grasping that swallowing has been performed has been proposed (see, for example, Patent Document 1 and Non-Patent Documents 1 to 3 below).

Patent Document 1 and Non-Patent Document 1 disclose a device that detects the sound of the throat of a subject with a microphone and outputs the detected sound from a speaker. According to these devices, a user such as a nurse or a caregiver listens to the sound of the subject's throat output from the speaker and determines whether or not swallowing is properly performed based on the subtle difference in sound. You can judge.

In Non-Patent Document 2, a subject's swallowing sound is detected by acceleration transducers arranged at a plurality of locations in the throat, and based on the acoustic signal of the detected sound, a swallowing sound production site and a swallowing sound production site and A method for elucidating the acoustic characteristics at each production site is disclosed.

Japanese Unexamined Patent Publication No. 2003-111748

"Cum checker", [online], Reserveia Co., Ltd., [Search on May 28, 2016], Internet <URL: http: // www. Aspiration pneumonia.com/> Yuji Nakayama, 5 outsiders, "Study of swallowing sound production site and acoustic characteristics", Showa Dental Society Magazine, Showa University Showa Dental Society, 2006, Vol. 26, p.163-174

By the way, the devices described in Patent Document 1 and Non-Patent Document 1 have only a function of simply outputting the sound of the throat of the subject from the speaker. Further, in the method described in Non-Patent Document 2, a sound that is known in advance to be the sound of the subject when swallowing is used.

However, since the actual sound of the throat includes not only the sound of swallowing but also the sound of chewing and speaking, it is not easy to distinguish the sound of swallowing from the sound of the throat. .. Therefore, it may be difficult to confirm whether or not swallowing has been performed.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a swallowing information presenting device that allows a user to easily confirm that swallowing has been performed.

The swallowing information presenting device of the present invention is a signal acquisition unit that acquires the throat sound, which is the sound of the subject's throat, as an acoustic signal, and a period during which the subject swallows by extracting a feature amount from the acoustic signal. A detection unit that detects the swallowing section based on the feature amount, a generation unit that converts an acoustic signal in the swallowing section into a signal for perceptual output when the detection unit detects the swallowing section, and a generation unit. It includes an output unit that outputs a signal converted by.

In this swallowing information presenting device, a feature amount is extracted from the acoustic signal of the subject's throat sound, and the swallowing section is detected based on the feature amount. Therefore, the swallowing section can be automatically identified. Further, when the swallowing section is detected, the acoustic signal in the swallowing section is converted into a signal for perceptual output, and the converted signal is perceptibly output. Therefore, it is possible to more reliably notify a user such as a nurse or a caregiver that the subject has swallowed. Therefore, the user can easily confirm that swallowing has been performed.

In the swallowing information presenting device of the present invention, the generation unit converts the acoustic signal in the swallowing section into a first reproduction signal for outputting as sound when the detection unit detects the swallowing section, and the output unit generates the first reproduction signal. , The first reproduction signal may be output as a swallowing sound. In this case, a user such as a nurse or a caregiver can easily confirm that swallowing has been performed by listening to the swallowing sound even when, for example, other work is being performed. ..

In this swallowing information presenting device, the generation unit may modulate the first reproduction signal into a modulated signal, and the output unit may output the modulated signal as a swallowing sound. In this case, since the swallowing sound can be more easily heard and output, the user can more easily confirm that the swallowing has been performed.

This swallowing information presenting device further includes a reproduction unit that converts an acoustic signal during a period including at least a part of the period excluding the swallowing section into a second reproduction signal for outputting as sound, and the output unit is a second reproduction unit. The signal may be output as a sore throat. In this case, the throat sound continues to be output even when the detection unit cannot detect the swallowing section or when the detection unit erroneously detects a period other than the swallowing section as a swallowing section. ing. Therefore, a user such as a nurse or a caregiver can distinguish the swallowing sound from the throat sound and confirm that the swallowing has been performed.

In this swallowing information presenting device, the generation unit or the reproduction unit may include a function capable of adjusting at least one of the intensity of the swallowing sound and the intensity of the throat sound. In this case, since the balance between the intensity of the swallowing sound and the intensity of the throat sound can be adjusted so that the swallowing sound becomes easier to hear, the user can more easily confirm that the swallowing has been performed.

In this swallowing information presenting device, the generation unit converts the acoustic signal in the swallowing section into a signal to be output as a detection sound stored in advance when the detection unit detects the swallowing section, and the output unit. May output the signal converted by the generation unit as the detection sound. In this case, since the detection sound that is easier to hear is output instead of the swallowing sound, the user can more easily confirm that the swallowing has been performed.

This swallowing information presenting device further includes a reproduction unit that converts an acoustic signal during a period including at least a part of the period excluding the swallowing section into a second reproduction signal for outputting as sound, and the output unit is a second reproduction unit. The signal may be output as a sore throat. In this case, the throat sound continues to be output even when the detection unit cannot detect the swallowing section or when the detection unit erroneously detects a period other than the swallowing section as a swallowing section. ing. Therefore, a user such as a nurse or a caregiver can distinguish the swallowing sound from the throat sound and confirm that the swallowing has been performed.

In this swallowing information presenting device, the generation unit or the reproduction unit may include a function capable of adjusting at least one of the intensity of the detected sound and the intensity of the throat sound. In this case, since the balance between the intensity of the detected sound and the intensity of the throat sound can be adjusted so that the detected sound can be heard more easily, the user can more easily confirm that swallowing has been performed.

This swallowing information presenting device further includes a reproduction unit that converts an acoustic signal in a period excluding the swallowing section into a second reproduction signal for outputting as sound, and the output unit outputs the second reproduction signal as a throat sound. You may. In this case, the throat sound continues to be output even when the detection unit cannot detect the swallowing section. Therefore, a user such as a nurse or a caregiver can distinguish the swallowing sound from the throat sound and confirm that the swallowing has been performed.

This swallowing information presenting device further includes a reproduction unit that converts an acoustic signal in a part of the period excluding the swallowing section and the swallowing section into a second reproduction signal for outputting as sound, and the output unit is a second reproduction signal. May be output as a throat sound. In this case, the throat sound continues to be output even when the detection unit cannot detect the swallowing section or when the detection unit erroneously detects a period other than the swallowing section as a swallowing section. ing. Therefore, a user such as a nurse or a caregiver can distinguish the swallowing sound from the throat sound and confirm that the swallowing has been performed.

In this swallowing information presenting device, the generation unit converts the acoustic signal in the swallowing section into a signal for outputting as light when the detection unit detects the swallowing section, and the output unit is converted by the generation unit. The signal may be output as light. In this case, a user such as a nurse or a caregiver can easily confirm that swallowing has been performed by seeing the light output by the output unit even when performing other work, for example. You can check.

In this swallowing information presenting device, the generation unit converts the acoustic signal in the swallowing section into a signal for outputting as vibration when the detection unit detects the swallowing section, and the output unit is converted by the generation unit. The signal may be output as vibration. In this case, a user such as a nurse or a caregiver can easily feel that the swallowing has been performed by feeling the vibration output by the output unit even when performing other work, for example. You can check.

According to the present invention, it is possible for the user to easily confirm that swallowing has been performed.

It is a block diagram which shows the swallowing information presenting apparatus which concerns on one Embodiment. It is a block diagram which shows the hardware structure of the swallowing information presenting apparatus. It is a flowchart which shows the presentation processing of the swallowing information by a swallowing information presenting apparatus.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a swallowing information presenting device according to an embodiment. The swallowing information presenting device 1 shown in FIG. 1 outputs (reproduces) the swallowing sound, which is the sound of the subject's throat (throat sound) during the period (swallowing section) during which the subject swallows, and thereby the subject. It is a device that informs the surroundings that the swallowing has been performed. An example of a user who is expected to use the swallowing information presenting device 1 is a nurse or a caregiver in a medical or long-term care field. In this case, the target person is a person who receives nursing or long-term care from the user. The users and target persons are not limited to these persons.

As shown in FIG. 1, the swallowing information presenting device 1 communicates with the microphone 2 by wire or wirelessly. The microphone 2 detects the sound of the subject's throat and outputs it to the swallowing information presenting device 1 as an acoustic signal which is an electric signal. The microphone 2 is held in the subject's throat by, for example, a fixture fixed to the subject's neck. The mode in which the microphone 2 is held in the throat of the subject is not particularly limited.

Further, the swallowing information presenting device 1 communicates with the first speaker 3 and the second speaker 4 by wire or wirelessly. The swallowing information presenting device 1 outputs a signal for outputting as a swallowing sound to the first speaker 3, and the first speaker 3 outputs a swallowing sound. Further, the swallowing information presenting device 1 outputs a signal for outputting as a throat sound to the second speaker 4, and the second speaker 4 outputs a throat sound.

FIG. 2 is a block diagram showing a hardware configuration of the swallowing information presenting device. As shown in the figure, the swallowing information presenting device 1 is physically a CPU 101, a RAM 102 and a ROM 103 which are main storage devices, an input device 104 which is an input device such as an input key and a touch sensor, a touch panel display, and a liquid crystal display. It is configured as a computer system (information processor) including an output device 105 such as, and an auxiliary storage device 106 such as a semiconductor memory. The processing function described later of the swallowing information presenting device 1 is to load one or a plurality of predetermined computer software on the hardware such as the CPU 101 and the RAM 102 shown in FIG. 2, and the input device 104 is controlled by the CPU 101. It is realized by operating the output device 105 and reading and writing data in the RAM 102 and the auxiliary storage device 106.

Returning to FIG. 1, the details of the functional components included in the swallowing information presenting device 1 will be described. The swallowing information presenting device 1 includes a signal acquisition unit 10, a detection unit 11, a generation unit 12, a reproduction unit 13, a delay unit 14, and an output unit 15 as functional components.

The signal acquisition unit 10 acquires the throat sound output by the microphone 2 as an acoustic signal. The signal acquisition unit 10 has an A / D conversion unit for A / D conversion of the acquired acoustic signal. The configuration of the A / D conversion unit is not particularly limited. The signal acquisition unit 10 A / D-converts the acquired acoustic signal and outputs it to the detection unit 11.

The detection unit 11 extracts a feature amount from the acoustic signal and detects a swallowing section based on the extracted feature amount. The detection unit 11 includes a sound section detection unit 20, a feature amount extraction unit 21, and an identification unit 22.

The sound section detection unit 20 determines a section (period) including a sound signal by VAD (Voice Activity Detection: sound section detection). In the present embodiment, the sound section detection unit 20 specifically detects a period in which the power (for example, logarithmic power) in the acoustic signal of the throat sound is equal to or higher than a predetermined value, and sets each detected period as a sound event. Is determined. Here, the sound event means an event that generates a sound, and examples thereof include swallowing, chewing, utterance, and the like by a subject, and environmental sounds. The specific VAD method that can be used in the sound section detection unit 20 is not limited to the above.

The feature amount extraction unit 21 extracts the feature amount in each sound event detected by the sound section detection unit 20. As the feature amount, for example, MFCC (Mel-Frequency Cepstrum Coefficients) and phase information of an acoustic signal can be used. The phase information of the acoustic signal is information regarding the difference in the phase of the waveform of the acoustic signal. When phase information is used, the acoustic signal is treated as a waveform that differs depending on the phase even if there is no difference in the power spectrum. Regarding the phase information used in the feature extraction unit 21, "Wang Ryu Mark, 3 outsiders," Evaluation of speaker identification / collation method using phase information ", IPSJ Research Report, 2008-SLP- 74 (30), p.173-178 ”and others.

The identification unit 22 is a discriminator, and identifies the label corresponding to the sound event based on the feature amount in each sound event extracted by the feature amount extraction unit 21. Here, the label is the cause of generating a sound in each sound event. The label includes at least swallowing by the subject, and may also include chewing, utterance, etc. by the subject. The sound section of the sound event identified by the identification unit 22 as the corresponding label is swallowing is detected as the swallowing section. In the present embodiment, the identification unit 22 is configured by using an SVM (Support Vector Machine).

When the detection unit 11 detects the swallowing section, the generation unit 12 converts the acoustic signal in the swallowing section into a signal for perceptually outputting the sound signal. Specifically, the generation unit 12 converts the acoustic signal in the swallowing section into a first reproduction signal for outputting as a sound when the detection unit 11 detects the swallowing section.

Here, the generation unit 12 modulates the first reproduction signal into a modulation signal for outputting as a swallowing sound. "Modulation" means changing the waveform of the first reproduction signal by changing at least one of the amplitude, frequency, and phase of the first reproduction signal. The "swallowing sound" is an output of an acoustic signal in the swallowing section as a perceptible sound. The generation unit 12 outputs the modulated signal to the output unit 15. It should be noted that the generation unit 12 takes a predetermined time to process the first reproduction signal to be modulated into a modulated signal and output the modulated signal to the output unit 15.

Further, the generation unit 12 includes an adjustment unit 23 having a function of adjusting the intensity of the swallowing sound output by the output unit 15. The output unit 15 outputs the swallowing sound and the throat sound will be described later.

The reproduction unit 13 is a second for outputting an acoustic signal as a throat sound regardless of whether or not the detection unit 11 has detected the swallowing section (that is, a part of the period excluding the swallowing section and the swallowing section). Convert to a playback signal. The "throat sound" is an output of an acoustic signal as a perceptible sound during a period not limited to the swallowing section. The reproduction unit 13 outputs the second reproduction signal to the delay unit 14.

The delay unit 14 is composed of a delay circuit and delays the second reproduction signal. The delay unit 14 outputs the delayed second reproduction signal to the output unit 15. The time required for the delay unit 14 to delay the second reproduction signal coincides with the time required for the generation unit 12 to process the first reproduction signal into a modulated signal and output the modulated signal to the output unit 15. It is set.

The output unit 15 has a D / A conversion unit for D / A conversion of the modulated signal and the second reproduction signal. The configuration of the D / A conversion unit is not particularly limited. The output unit 15 D / A-converts the modulated signal input from the generation unit 12 and outputs it to the first speaker 3. Further, the output unit 15 D / A-converts the second reproduction signal input from the delay unit 14 and outputs the second reproduction signal to the second speaker 4. As a result, the output unit 15 outputs the modulated signal as a swallowing sound via the first speaker 3 and outputs the second reproduced signal as a throat sound via the second speaker. In other words, the output unit 15 outputs the swallowing sound and the throat sound in stereo via the first speaker 3 and the second speaker 4. The output unit 15 may add the swallowing sound and the throat sound in a time-synchronized manner and output the sound via both the first speaker 3 and the second speaker 4.

The adjustment unit 23 of the generation unit 12 described above has a function of adjusting the intensity of the swallowing sound output by the output unit 15. That is, the adjusting unit 23 adjusts the balance between the intensity of the swallowing sound and the intensity of the throat sound. For example, the adjustment unit 23 is realized by a volume adjustment button, and the balance is adjusted by the button. Alternatively, the adjusting unit 23 may include an input means realized by software and may be configured to adjust the balance by the input means, or may adjust the balance by a parameter acquired from the outside by communication or the like. It may be configured to adjust. The adjusting unit 23 may have a function of adjusting both the intensity of the swallowing sound and the intensity of the throat sound. In that case, the adjusting unit 23 may have either the swallowing sound or the throat sound. The strength of one may be increased and the strength of either one may be decreased.

Subsequently, the operation of the swallowing information presenting device 1 described above will be described with reference to FIG. FIG. 3 is a flowchart showing a swallowing information presentation process by the swallowing information presenting device.

When the instruction to start the swallowing information presentation process is received, the process shown in the flowchart of FIG. 3 is started. First, the acoustic signal of the throat sound output by the signal acquisition unit 10 is acquired by the microphone 2 (step S1). Then, the acoustic signal acquired by the signal acquisition unit 10 is output to the sound section detection unit 20 of the detection unit 11.

Next, the sound section detection unit 20 performs VAD on the acoustic signal. Specifically, the sound section detection unit 20 detects a period in which the power (for example, logarithmic power) in the acoustic signal of the throat sound is equal to or higher than a predetermined value, and each detected period is determined to be a sound event. (Step S2). Then, the feature amount extraction unit 21 extracts the feature amount (for example, MFCC and the phase information of the acoustic signal in each sound event) in each sound event detected by the sound section detection unit 20 (step S3). .. Further, the identification unit 22 identifies the sound event corresponding to swallowing by the subject based on the feature amount in each sound event extracted by the feature amount extraction unit 21 (step S4). The sound section of the sound event corresponding to the swallowing of the subject identified in this way is determined to be the swallowing section.

Next, when the swallowing section is detected by the detection unit 11, the generation unit 12 converts the acoustic signal in the swallowing section into the first reproduction signal (step S5), and the first reproduction signal is modulated into the modulation signal. (Step S6). The modulated signal is output from the generation unit 12 to the output unit 15.

Here, in synchronization with the processing of steps S5 and S6, the reproduction unit 13 converts the acoustic signal into a second reproduction signal (step S7), and outputs the second reproduction signal to the delay unit 14. The second reproduction signal is delayed in the delay unit 14 (step S8). The delayed second reproduction signal is output from the delay unit 14 to the output unit 15.

Then, the modulated signal is output by the output unit 15 as a swallowing sound, and the second reproduction signal is output by the output unit 15 as a throat sound. At this time, the swallowing sound and the throat sound are output in time-synchronous addition or stereo via the first speaker 3 and the second speaker 4 (step S9).

At this time, the throat sound is output by the second speaker 4 while the swallowing sound is output by the first speaker 3. Therefore, a user such as a nurse or a caregiver swallows from the throat sound output by the second speaker 4 while confirming that the subject has swallowed by the swallowing sound output by the first speaker 3. By distinguishing the sounds, the detection unit 11 correctly detected the swallowing section (that is, not when the detection unit 11 could not detect the swallowing section, and the detection unit 11 detects the period other than the swallowing section. It is not the case that it is mistakenly detected as a swallowing section).

The processes of steps S1 to S9 are repeated until an instruction to end the swallowing information presentation process is received.

As described above, in the swallowing information presenting device 1, the feature amount is extracted from the acoustic signal of the throat sound of the subject, and the swallowing section is detected based on the feature amount. Therefore, the swallowing section can be automatically identified. Further, when the swallowing section is detected, the acoustic signal in the swallowing section is converted into a signal for being perceptibly output as sound, and the converted signal is output as sound. For this reason, users such as nurses and caregivers can use their ears to reliably confirm that the subject has swallowed, even while working with their hands or eyes. it can. Therefore, even when the user is working, the user can easily confirm that swallowing has been performed.

Further, in the swallowing information presenting device 1, the generation unit 12 converts the acoustic signal in the swallowing section into a first reproduction signal for outputting as sound when the detection unit 11 detects the swallowing section, and outputs the sound signal. The unit 15 outputs the first reproduction signal as a swallowing sound. Therefore, a user such as a nurse or a caregiver can easily confirm that swallowing has been performed by listening to the swallowing sound even when, for example, other work is being performed. ..

Further, in the swallowing information presenting device 1, the generation unit 12 modulates the first reproduction signal into a modulated signal, and the output unit 15 outputs the modulated signal as a swallowing sound. Therefore, since the swallowing sound can be more easily heard and output, the user can more easily confirm that the swallowing has been performed.

Further, the swallowing information presenting device 1 further includes a reproduction unit 13 that converts an acoustic signal during a period including at least a part of the period excluding the swallowing section into a second reproduction signal for outputting as sound, and the output unit 15 further includes a reproduction unit 13. , The second reproduction signal is output as a throat sound. Therefore, even when the detection unit 11 cannot detect the swallowing section, or when the detection unit 11 erroneously detects a period other than the swallowing section as the swallowing section, the throat sound is output. It continues to be done. Therefore, a user such as a nurse or a caregiver can distinguish the swallowing sound from the throat sound and confirm that the swallowing has been performed.

Further, in the swallowing information presenting device 1, the generation unit 12 includes a function of adjusting the intensity of the swallowing sound. Therefore, the balance between the intensity of the swallowing sound and the intensity of the throat sound can be adjusted so that the swallowing sound can be heard more easily, so that the user can more easily confirm that the swallowing has been performed.

Further, the swallowing information presenting device 1 further includes a reproduction unit 13 that converts an acoustic signal in a part of the period excluding the swallowing section and the swallowing section into a second reproduction signal for outputting as sound, and the output unit 15 further includes a reproduction unit 13. The second reproduction signal is output as a throat sound. Therefore, even when the detection unit 11 cannot detect the swallowing section, or when the detection unit 11 erroneously detects a period other than the swallowing section as the swallowing section, the throat sound is output. It continues to be done. Therefore, a user such as a nurse or a caregiver can distinguish the swallowing sound from the throat sound and confirm that the swallowing has been performed.

The present invention is not limited to the above-described embodiments. For example, in the above embodiment, the generation unit 12 modulates the first reproduction signal into a modulated signal and outputs it to the output unit 15, and the output unit 15 outputs the modulated signal as a swallowing sound. However, the generation unit 12 may output the first reproduction signal to the output unit 15 without modulating it into a modulation signal, and the output unit may output the first reproduction signal as a swallowing sound.

Further, in the above embodiment, the reproduction unit 13 converts the acoustic signal into a second reproduction signal for outputting as sound, and the output unit 15 outputs the second reproduction signal as a sore throat sound. However, the swallowing information presenting device 1 does not include the reproduction unit 13, and the output unit 15 does not have to output the throat sound.

Further, the generation unit 12 converts the acoustic signal in the swallowing section into a signal for outputting as a detection sound stored in advance when the detection unit 11 detects the swallowing section, and the output unit 15 converts the acoustic signal into a signal for outputting as a detection sound stored in advance. The signal converted by the generation unit 12 may be output as a detection sound. In this case, since the detection sound that is easier to hear is output instead of the swallowing sound, the user can more easily confirm that the swallowing has been performed.

The detected sound is not particularly limited. For example, the detection sound may be a typical pre-recorded swallowing sound, or may be a sound (onomatopoeia) such as "gokkun" by an artificially created synthetic sound, or a default sound such as a chime. It may be a sound.

Further, the length of the detected sound may be changed according to the length of the swallowing section. For example, the length of the detection sound may be changed so as to be the same length as the length of the swallowing section (for example, the length of time required from the start to the completion of swallowing). For that purpose, the length of the detected sound may be changed by adjusting the reproduction speed of the detected sound, or the detected sound may be repeatedly output until the length becomes equal to the length of the swallowing section. ..

Here, when the output unit 15 outputs the detection sound, the adjustment unit 23 may have a function of adjusting the intensity of the detection sound, or the adjustment unit 23 may have a function of adjusting the intensity of the detection sound. It may have a function that can adjust both the intensity of the throat sound. In this case, since the balance between the intensity of the detected sound and the intensity of the throat sound can be adjusted so that the detected sound can be heard more easily, the user can more easily confirm that swallowing has been performed.

Further, the adjusting unit 23 may be included in the reproducing unit 13 and may have a function of adjusting the intensity of the throat sound output by the output unit 15. Alternatively, the adjusting unit 23 may be included in the reproducing unit 13 and may have a function of adjusting both the intensity of the swallowing sound (or the detected sound) and the intensity of the throat sound.

Further, in the above embodiment, the reproduction unit 13 converts an acoustic signal in a part of the period excluding the swallowing section and the swallowing section into a second reproduction signal for outputting as a sound, and converts the second reproduction signal into a throat sound. Output as. However, the reproduction unit 13 converts the acoustic signal in the period (only) excluding the swallowing section into a second reproduction signal for outputting as sound, and the output unit 15 outputs the second reproduction signal as a throat sound. May be good. In this case, even when the detection unit 11 cannot detect the swallowing section, the throat sound continues to be output. Therefore, a user such as a nurse or a caregiver can distinguish the swallowing sound from the throat sound and confirm that the swallowing has been performed.

Further, in the above embodiment, the phase information of the MFCC and the acoustic signal at each sound event is used as the feature amount. However, as the feature amount, in addition to these MFCC and phase information, the average value of energy in each sound event, the continuation length of each sound event, etc. may be used, and the environmental sound is acquired separately from the throat sound. Where possible, the average value of the power and power ratio between the throat sound and the environmental sound in each sound event, the MFCC and energy of the environmental sound, and the like may be further used. The phase information may not be used as the feature amount.

Further, in the above embodiment, the identification unit 22 is configured by using an SVM. However, the identification unit 22 is not limited to SVM, and is configured by using, for example, a decision tree, GMM (Gaussian Mixture Model), AdaBoost (AdaptiveBoosting), HMM (Hidden Markov Model), or the like. It may have been.

Further, in the above embodiment, the swallowing information presenting device 1 communicates with the first speaker 3 and the second speaker 4 by wire or wirelessly, and outputs the swallowing sound and the throat sound by time-synchronized addition or stereo. doing. However, the swallowing information presenting device 1 may communicate with one speaker by wire or wirelessly, and output the swallowing sound and the throat sound in a time-synchronized addition.

Further, in the above embodiment, the signal acquisition unit 10 has an A / D conversion unit, and the output unit 15 has a D / A conversion unit. However, the signal acquisition unit 10 does not have to have an A / D conversion unit, and the output unit 15 does not have to have a D / A conversion unit.

Further, in the above embodiment, the generation unit 12 converts the acoustic signal in the swallowing section into a signal for perceptually outputting the sound as a sound when the detection unit 11 detects the swallowing section. However, the perceptible output is not limited to sound, and is not particularly limited as long as it is an output that can be perceived by the five senses of the user.

For example, the generation unit 12 converts the acoustic signal in the swallowing section into a signal for outputting as light when the detection unit 11 detects the swallowing section, and the output unit 15 is converted by the generation unit 12. The signal may be output as light. When the acoustic signal in the swallowing section is converted into light, the swallowing information presenting device 1 may be connected to a light emitting device such as an LED instead of the first speaker 3 and the second speaker 4. In this case, a user such as a nurse or a caregiver can easily confirm that swallowing has been performed by looking at the LED light even when performing other work, for example. it can.

Alternatively, the generation unit 12 converts the acoustic signal in the swallowing section into a signal for outputting as vibration when the detection unit 11 detects the swallowing section, and the output unit 15 is converted by the generation unit 12. The signal may be output as vibration. When the acoustic signal in the swallowing section is converted into vibration, the swallowing information presenting device 1 may be connected to a vibration generating device such as a vibrator instead of the first speaker 3 and the second speaker 4. .. In this case, a user such as a nurse or a caregiver can easily confirm that swallowing has been performed by feeling the vibration of the vibrator even when performing other work, for example. it can.

1 ... swallowing information presenting device, 2 ... microphone, 3 ... first speaker, 4 ... second speaker, 10 ... signal acquisition unit, 11 ... detection unit, 12 ... generation unit, 13 ... reproduction unit, 14 ... delay unit, 15 ... Output unit, 20 ... Sound section detection unit, 21 ... Feature amount extraction unit, 22 ... Identification unit, 23 ... Adjustment unit.

Claims (9)

A signal acquisition unit that acquires the throat sound, which is the sound of the subject's throat, as an acoustic signal,
A detection unit that extracts a feature amount from the acoustic signal and detects a swallowing section, which is the period during which the subject swallows, based on the feature amount.
When the detection unit detects the swallowing section, a generation unit that converts the acoustic signal in the swallowing section into a signal for perceptual output, and a generation unit.
An output unit that outputs a signal converted by the generation unit is provided .
When the detection unit detects the swallowing section, the generation unit converts the acoustic signal in the swallowing section into a first reproduction signal for outputting as sound.
The output unit is a swallowing information presenting device that outputs the first reproduction signal as a swallowing sound . The generator modulates the first reproduction signal into a modulated signal,
The output unit outputs the modulated signal as said swallowing sound, swallowing information presentation apparatus according to claim 1. A reproduction unit for converting the acoustic signal into a second reproduction signal for outputting as sound in a period including at least a part of the period excluding the swallowing section is further provided.
The swallowing information presenting device according to claim 1 or 2 , wherein the output unit outputs the second reproduction signal as the throat sound. The swallowing information presenting device according to claim 3 , wherein the generation unit or the reproduction unit includes a function of adjusting at least one of the intensity of the swallowing sound and the intensity of the throat sound. A signal acquisition unit that acquires the throat sound, which is the sound of the subject's throat, as an acoustic signal,
A detection unit that extracts a feature amount from the acoustic signal and detects a swallowing section, which is the period during which the subject swallows, based on the feature amount.
When the detection unit detects the swallowing section, a generation unit that converts the acoustic signal in the swallowing section into a signal for perceptual output, and a generation unit.
An output unit that outputs the signal converted by the generation unit, and
A reproduction unit for converting the acoustic signal into a second reproduction signal for output as sound in a period including at least a part of the period excluding the swallowing section is provided.
When the detection unit detects the swallowing section, the generation unit converts the acoustic signal in the swallowing section into a signal for outputting as a detection sound stored in advance.
The output unit outputs the signal converted by the generation unit as the detection sound .
The output unit is a swallowing information presenting device that outputs the second reproduction signal as the throat sound . The swallowing information presenting device according to claim 5 , wherein the generating unit or the reproducing unit includes a function of adjusting at least one of the intensity of the detected sound and the intensity of the throat sound. A reproduction unit for converting the acoustic signal in a period excluding the swallowing section into a second reproduction signal for outputting as sound is further provided.
The swallowing information presenting device according to any one of claims 3 to 6 , wherein the output unit outputs the second reproduction signal as the throat sound. A reproduction unit for converting a part of the period excluding the swallowing section and the acoustic signal in the swallowing section into a second reproduction signal for outputting as sound is further provided.
The swallowing information presenting device according to any one of claims 3 to 6 , wherein the output unit outputs the second reproduction signal as the throat sound. The first step of acquiring the throat sound, which is the sound of the subject's throat, as an acoustic signal,
The second step of detecting the sound section based on the acoustic signal and
The third step of extracting the feature amount of the acoustic signal in the sound section, and
A fourth step of detecting a swallowing section, which is a sound section swallowed by the subject, based on the feature amount in the sound section.
When the swallowing section is detected in the fourth step, the fifth step of converting the acoustic signal in the swallowing section into a first reproduction signal for outputting as sound, and
The sixth step of modulating the first reproduced signal into a modulated signal, and
The seventh step of converting the acoustic signal into a second reproduction signal for outputting as sound, and
The eighth step of delaying the second reproduction signal and
The ninth step of outputting the modulated signal as a swallowing sound and the second reproduction signal as a throat sound,
A swallowing information presentation device that performs.

Images