JP2014083279A - Mastication frequency calculation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To count a mastication frequency without mounting an instrument on a user.SOLUTION: Sound data obtained by collecting sound while a user is taking a meal and weight data obtained by measuring the change of the weight of the meal during taking the meal are input. A section where mastication sound is recorded is extracted from sound data. A section during a mastication action is extracted on the basis of the weight change of the weight data. Then, a mastication frequency is calculated. Consequently, a part erroneously recognized as mastication sound can be removed based on the weight change of the meal. Thus, the mastication frequency is correctly calculated based on the sound data collected by a microphone 21 arranged on a table without mounting an instrument on the user.

Description

  The present invention relates to a technique for counting the number of chewing times during a meal.

  Appropriate chewing exercises are said to be less likely to cause tooth decay due to a large amount of saliva, and less likely to cause diseases such as gastric cancer because food is often chewed and the digestive burden on the internal organs is reduced. You can live a healthier life by grasping your own number of chewing times and eating at the appropriate number of chewing times. In recent years, an apparatus that counts the number of mastications by attaching an instrument to the outer ear has also been proposed (for example, Patent Documents 1 and 2).

JP 2008-048791 A JP 2011-010791 A

  However, the techniques of Patent Documents 1 and 2 have a problem that it is troublesome because an apparatus must be attached to the face or the like during a meal. In addition, there is a problem that speech and tongue movement are counted in the number of chewing times.

  The present invention has been made in view of the above, and an object of the present invention is to count the number of mastications without causing a user to wear a device.

  The mastication frequency calculating apparatus according to the first aspect of the present invention is an input means for inputting voice data obtained by collecting sounds during a meal of a user and weight data obtained by measuring a weight change of a dish during the meal, and mastication from the voice data. Mastication sound extraction means for extracting a section in which sound is recorded, motion detection means for estimating a section during mastication operation from a weight change of the weight data, a section in which the mastication sound of the voice data is recorded, and the mastication And mastication number calculating means for obtaining the number of mastication times of the user from the operating section.

  In the mastication frequency calculation device, the mastication frequency calculation means supplements the mastication frequency based on the mastication speed of the user for each food that has been registered in advance for a portion where it was difficult to obtain a mastication sound in the interval during the mastication operation. It is characterized by that.

  In the above chewing frequency calculation device, the voice data is collected by a microphone installed without being attached to a user's body, and the weight data is measured by a plate-type weight scale on which a dish is placed. .

  According to the present invention, it is possible to count the number of chewing times without causing the user to wear a device.

It is a whole block diagram including the mastication frequency calculation apparatus in this Embodiment. It is a sequence diagram which shows the flow of the process which calculates the number of chewing. It is a flowchart which shows the flow of the process which a weight measurement sound collector collects weight data and audio | voice data, and transmits to a chewing frequency calculation apparatus. It is a flowchart which shows the flow of the process in which a chewing number calculation apparatus calculates the number of mastication. It is a figure which shows the audio | voice data and weight data which a chewing frequency calculation apparatus receives. It is a figure explaining a mode that the sound before mastication operation | movement is removed as misidentification information based on weight data. It is a figure which shows typically that a chewing speed changes with the hardness of the foodstuff of cooking.

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

  FIG. 1 is an overall configuration diagram including a mastication number calculating apparatus according to the present embodiment. The mastication frequency calculation device 1 according to the present embodiment receives the audio data including the user's mastication sound and the weight data obtained by measuring the weight change of the food from the weight measurement sound collecting device 2 arranged in the user's home, and calculates the mastication frequency. It is a device to calculate. The weight measurement sound collection device 2 is installed on a user's table, collects sounds during meals, measures changes in the weight of food, and performs sound through a home GW (gateway) 4 connected to the network 100. Send data and weight data. In addition, the user can access the Web server 18 via the network 100 with the browsing terminal 3 such as a portable terminal or a personal computer, and can confirm the analysis result of the mastication number calculating device 1. Hereinafter, each device will be described.

  The mastication frequency calculating device 1 includes a receiving unit 11, a removing unit 12, an operation detecting unit 13, a complementing unit 14, an analyzing unit 15, and a dish information accumulating unit 16, and stores the analysis result in the mastication information accumulating unit 17. Each unit included in the mastication number calculating device 1 may be configured by a computer including an arithmetic processing device, a storage device, and the like, and the processing of each unit may be executed by a program. This program is stored in a storage device included in the mastication number calculating device 1, and can be recorded on a recording medium such as a magnetic disk, an optical disk, or a semiconductor memory, or provided through a network.

  The receiving unit 11 receives and holds audio data and weight data from the weight measurement sound collector 2.

  The removal unit 12 has a mastication sound pattern, matches the sound data received by the reception unit 11 with the mastication sound pattern, extracts a mastication sound segment in the sound data, and converts the mastication sound pattern to the mastication sound pattern. Unmatched parts are removed as speech sounds.

  The motion detection unit 13 refers to the weight data received by the reception unit 11, estimates that the user is performing a mastication operation immediately after the weight change, and generates a mastication sound in a section where the mastication operation is not performed in the audio data. The removed part is removed as misidentification information. And the number of mastication is calculated | required from the audio | voice data after removing the unnecessary part by the removal part 12 and the operation | movement detection part 13. FIG.

  The complement unit 14 supplements the number of mastications based on the user's mastication speed for the portion where it was difficult to obtain the mastication sound in the section estimated to be during the mastication operation, and provides the mastication information storage unit 17 with the mastication information such as the obtained number of mastications. Send.

  The analysis unit 15 receives the cooking image data and the cooking recipe from the user, determines the ingredients contained in the dish to determine the hardness of the ingredients, and determines the chewing speed of the user with respect to the determined hardness of the dishes. 16 and transmitted to the complementing unit 14.

  The cooking information storage unit 16 stores a plurality of types of mastication speeds for the hardness of the ingredients for each user. Information on mastication speed for each user is received from the user in advance and registered in advance.

  The mastication information storage unit 17 receives mastication information such as the number of mastication times, a mastication time, and a meal time from the mastication frequency calculation device 1, and accumulates the mastication information for each user.

  In response to an inquiry from the user, the Web server 18 reads out the mastication information from the mastication information storage unit 17 and presents it to the user. As a presentation method, there are methods such as a graph and a table.

  The weight measurement sound collecting apparatus 2 includes a microphone 21 and a weight scale 22 and is installed on a dining table. In addition, although not shown in figure, the transmission means which transmits data to the mastication frequency calculation apparatus 1 via the home | home GW4 and the network 100 is also provided. The weight measurement sound collector 2 starts sound collection and weight measurement when the microphone 21 and the weight scale 22 are powered on. When there is no change in the weight measured by the weighing scale 22 for a predetermined time, it is determined that the meal has ended, and the sound data collected during the meal and the weight data are transmitted to the mastication number calculating device 1 to turn on the power. Turn off.

  The microphone 21 is a highly sensitive sound collecting device that collects sounds during meals including a user's mastication sound. The microphone 21 only needs to be able to collect sounds during meals, so it need not be attached to the user's face or the like, and may be placed on the table or hung from the ceiling.

  The weigh scale 22 measures a change in the weight of the food during the meal. The weighing scale 22 is installed on the table in a plate shape to measure the weight of the dish, and the dish is arranged on the plate-shaped weighing scale 22.

  The browsing terminal 3 accesses the Web server 18 and acquires and displays the calculation result of the mastication number calculating device 1. In addition, in order to notify the cooking menu 1 to the cooking menu, the browsing terminal 3 may be equipped with a camera to take a picture of the cooking and transmit it to the cooking mute 1.

  The home GW 4 connects the weight measurement sound collector 2 and the browsing terminal 3 connected to the home LAN to the external network 100.

  Next, an operation for calculating the number of chewing times will be described.

  First, the overall processing flow including the mastication number calculating device 1 and the weight measurement sound collecting device 2 will be described.

  FIG. 2 is a sequence diagram showing a flow of processing for calculating the number of chewing times.

  First, the user notifies the chewing number calculation device 1 of information on the food to eat (step S11).

  When the user turns on the power of the weight measurement sound collection device 2 and starts eating, the weight measurement sound collection device 2 starts collecting the sound of mastication and measuring the weight of the dish (step S12).

  When the user's meal ends, the weight measurement sound collecting device 2 stops the measurement (step S13), and transmits the voice data and the weight data to the mastication number calculating device 1 (step S14).

  Then, the mastication frequency calculation device 1 analyzes the received voice data and weight data to calculate the mastication frequency (step S15), and transmits mastication information including the calculated mastication frequency to the mastication information storage unit 17 (step S16). ).

  The mastication information storage unit 17 registers the received mastication information (step S17).

  When the user accesses the Web server 18 with the browsing terminal 3, the Web server 18 reads the user's mastication information from the mastication information storage unit 17 and displays it on the browsing terminal 3, thereby browsing the mastication information (step S18).

  Next, the operation of the weight measurement sound collector 2 will be described.

  FIG. 3 is a flowchart showing a flow of processing in which the weight measurement sound collecting device 2 collects voice data and weight data and transmits them to the mastication number calculating device 1.

  First, the data of the food that the user eats is transmitted to the mastication number calculating device 1 (step S21). In the present embodiment, the browsing terminal 3 transmits cooking image data, cooking recipes, and the like to the mastication number calculating device 1. Instead of the image data of the dish, information such as the name of the dish and ingredients used in the dish may be transmitted.

  When the user starts a meal, the microphone 21 collects and records the sound of the user during the meal, and the weight scale 22 measures and records a change in the weight of the food during the meal (step S22).

  Then, it is determined whether or not the meal has ended based on the change in the weight of the dish (step S23). If it is determined that the meal has ended, the process proceeds to step S24. For example, when the weight of the dish does not change for a predetermined time or more, it is determined that the meal is finished.

  When the meal ends, the weight measurement sound collecting device 2 transmits the sound data and the weight data to the mastication number calculating device 1 (step S24).

  Then, the operation | movement of the chewing frequency calculation apparatus 1 is demonstrated.

  FIG. 4 is a flowchart showing a flow of processing in which the mastication number calculating device 1 calculates the number of mastication times.

  The receiving unit 11 receives the audio data and the weight data from the weight measurement sound collector 2 and stores them in a storage means such as a memory (step S31).

  The remover 12 matches the received voice data with the mastication sound pattern and removes the utterance sound from the voice data (step S32).

  The motion detection unit 13 estimates a predetermined period after the weight of the dish is reduced from the received weight data as being during the mastication operation, and misidentifies the portion detected as a mastication sound outside the interval estimated as during the mastication operation (Step S33). Since the sections of the mastication sound in the voice data are specified by the processes of steps S32 and S33, the number of mastication times is obtained from the mastication sound and the specified voice data. Note that steps S32 and S33 may be performed in the reverse order.

  If there is a portion where it is difficult to obtain a sound in the section estimated to be chewing, the analysis unit 15 obtains the hardness of the food from the registered cooking data and determines the chewing speed of the user with respect to the hardness of the food. Acquired from the information storage unit 16, the complementing unit 14 complements the number of chewing times from the chewing speed and the time length of the part where it was difficult to obtain a sound (step S 34).

  Then, the mastication information such as the number of mastications and the mastication time obtained in the above steps S32 to S34 is transmitted to the mastication information storage unit 17 and registered (step S35).

  Next, processing of the removal unit 12, the motion detection unit 13, and the complementing unit 14 will be described.

  FIG. 5 shows data received by the mastication number calculating apparatus 1.

  FIG. 5A is a diagram showing audio data received by the mastication frequency calculating apparatus 1. As shown in the figure, the audio data received by the mastication frequency calculation device 1 includes not only mastication sounds but also speech sounds during meals and other sounds. Therefore, the removal unit 12 matches the sound pattern of the mastication sound with the received sound data, and removes speech sounds other than the mastication sound from the sound data.

  FIG.5 (b) is a figure which shows the weight data which the chewing frequency calculation apparatus 1 receives. The weight data received by the chewing frequency calculation device 1 is weight data obtained by measuring a change in the weight of the food during the meal, and the weight of the food decreases at the timing when the user takes the food with chopsticks or the like. Therefore, it is assumed that the motion detection unit 13 estimates that the mastication operation is being performed for a predetermined period after the weight is reduced, and the portion matched as the mastication sound before the mastication operation is misidentified.

  FIG. 6 is a diagram for explaining a state in which a sound before the mastication operation is removed as misidentification information based on the weight data. The motion detection unit 13 associates the time information of the voice data and the weight data, and determines that the part matched as the mastication sound with respect to the voice data before the mastication operation is misidentified and removes it from the voice data. Since the portion indicated by the symbol A in FIG. 6 is presumed to be before the mastication operation, it is determined that it has been misidentified as a mastication sound and is removed from the audio data. Note that the length of the predetermined period estimated as being during the mastication operation may be a method of registering in advance for each user or a method of confirming utterance with voice data after weight reduction.

  FIG. 7 is a diagram schematically showing that the chewing speed varies depending on the hardness of the ingredients of cooking. When the food is hard, the chewing speed is slow (the upper waveform in FIG. 7), and when the food is normal, the chewing speed is faster than the hard (the lower waveform in FIG. 7).

  In the section where it is estimated that the voice data is being chewed, the complementing unit 14 complements the number of chewing times based on the chewing speed corresponding to the hardness of the food. The user registers in advance the mastication speed corresponding to each of the hardness of the plurality of ingredients in the dish information storage unit 16 of the mastication number calculation device 1. For example, the chewing speed is registered for each of two patterns of a hard food and a normal food. The analysis unit 15 receives information on the food to eat before eating, determines the hardness of the food based on the received information on the food, and determines the chewing speed.

  In the present embodiment, the embodiment in which voice data and weight data are transmitted / received via the network 100 has been described. However, an apparatus in which the mastication number calculating device 1 and the weight measuring sound collecting device 2 are integrated can also be used.

  In addition, the weight measuring and collecting apparatus 2 transmits voice data and weight data at the end of a meal. However, the weight measurement sound collecting apparatus 2 may sequentially transmit the voice data and the weight data in real time or may be sequentially transmitted at a predetermined timing during the meal. Good.

  When calculating the number of chewing times of a plurality of users, an area ID for identifying the weight scale 22 is assigned, the user and the weight scale 22 are connected, weight data is transmitted together with the area ID, and a change in weight of each user's dish is measured. To detect.

  As described above, according to the present embodiment, voice data obtained by collecting sounds during a user's meal and weight data obtained by measuring changes in the weight of food during meals are input, and mastication sounds are recorded from the voice data. By extracting the sections that are being chewed and calculating the number of mastications from the weight change of the weight data, the part that is mistaken as a chewing sound can be removed based on the weight change of the dish. It is possible to calculate the number of mastications more accurately from the voice data collected by the microphone 21 installed on the table without wearing a tool to the user.

  According to the present embodiment, by calculating the user's mastication speed from the hardness of the ingredients of the cooking, and supplementing the number of mastications based on the user's mastication speed for the portion where the mastication sound was difficult to obtain during the mastication operation, It becomes possible to calculate the exact number of chewing times.

  As an application example of the present invention, an alert is sent to the user when the number of chewing times required at one time is less or the meal time is likely to end extremely quickly, or based on exercise log and meal data, It is possible to recommend dishes and ingredients that optimize the number of chewing times and meal times.

DESCRIPTION OF SYMBOLS 1 ... Mastication frequency calculation apparatus 11 ... Reception part 12 ... Removal part 13 ... Motion detection part 14 ... Complement part 15 ... Analysis part 16 ... Cooking information storage part 17 ... Mastication information storage part 18 ... Web server 2 ... Weight measurement sound collection apparatus 21 ... Microphone 22 ... Weigh scale 3 ... Reading terminal 4 ... Home GW
100 ... Network

Claims (3)

Input means for inputting voice data obtained by collecting sounds during the meal of the user and weight data obtained by measuring a change in the weight of the food during the meal;
Mastication sound extraction means for extracting a section in which mastication sound is recorded from the voice data;
An action detecting means for estimating a section during a mastication action from a change in weight of the weight data;
Mastication number calculating means for obtaining the number of mastication of the user from the section in which the mastication sound of the voice data is recorded and the section during the mastication operation;
A device for calculating the number of mastication times characterized by comprising:   The mastication number calculating means complements the number of mastications based on a user's mastication speed for each food that has been registered in advance for a portion where it was difficult to obtain a mastication sound in the section during the mastication operation. The mastication number calculating apparatus according to 1.   The sound data is collected by a microphone installed without being attached to a user's body, and the weight data is measured by a plate-shaped weight scale on which a dish is placed. Chewing frequency calculation device.

Images