Sound analysis system, sound analysis method and program

Abstract

The present invention relates to a sound analysis system, a sound analysis method and a storage medium, which can suppress the decrease in the detection accuracy of sound pressure and perform speech analysis with high accuracy. The sound analysis system comprises: a first and a second sound pressure acquisition unit, each of which is arranged on a device worn by a user, and is respectively arranged at different distances from the user's mouth when the user wears the device, and respectively acquires the sound pressure of the user's voice; a distance estimation unit, which estimates the distance between the first or second sound pressure acquisition unit and the user's mouth based on the sound pressure obtained by the first sound pressure acquisition unit and the sound pressure obtained by the second sound pressure acquisition unit; and a sound pressure correction unit, which calculates the difference between a reference value of the distance between the first or second sound pressure acquisition unit and the user's mouth and the distance estimated by the distance estimation unit, and corrects the sound pressure obtained by at least one of the first and second sound pressure acquisition units based on the difference.

Description

Technical Field

[0001] The present invention relates to a sound analysis system, a sound analysis method and a program. Background Art

[0002] A speech analysis system is disclosed in which a wearable terminal is suspended from the neck of a user and is provided with two sound pressure sensors disposed at different distances from the user's mouth (for example, see Japanese Patent No. 6191747). The sound pressure analysis system determines whether the sound pressure is generated by the user or the surrounding speech based on the sound pressure ratio obtained by each speech sensor. Summary of the invention

[0003] However, sometimes, due to reasons such as twisting of the suspension rope on which the voice sensor is installed, the distance between the user's mouth and each voice sensor changes, and the sound pressure obtained by each voice sensor also changes. In this case, the detection accuracy of the sound pressure decreases, and the accuracy of voice analysis may also decrease.

[0004] The present invention has been made to solve such a problem, and a main object of the present invention is to provide a speech analysis system, a speech analysis method, and a program that can suppress a decrease in the detection accuracy of sound pressure and perform speech analysis with high accuracy.

[0005] One solution of the present invention for achieving the above-mentioned object is a sound analysis system.

[0006] The first sound pressure acquisition unit and the second sound pressure acquisition unit are respectively arranged on the equipment worn by the user, and are respectively arranged at different distances from the user's mouth when the user wears the equipment, and respectively acquire the sound pressure of the user's voice;

[0007] a distance estimating unit that estimates a distance between the first sound pressure acquiring unit or the second sound pressure acquiring unit and a mouth of a user based on the sound pressure acquired by the first sound pressure acquiring unit and the sound pressure acquired by the second sound pressure acquiring unit; and

[0008] The sound pressure correction unit calculates the difference between a reference value of the distance between the first sound pressure acquisition unit or the second sound pressure acquisition unit and the user's mouth and the distance estimated by the distance estimation unit, and corrects the sound pressure acquired by at least one of the first sound pressure acquisition unit and the second sound pressure acquisition unit based on the difference.

[0009] In this embodiment, the distance estimation unit may estimate the distance between the first sound pressure acquisition unit or the second sound pressure acquisition unit and the user's mouth based on the sound pressure acquired by the first sound pressure acquisition unit and the second sound pressure acquisition unit and a distance correspondence mapping, function or learner representing the relationship between the sound pressure acquired by the first sound pressure acquisition unit and the second sound pressure acquisition unit and the distance between the first sound pressure acquisition unit or the second sound pressure acquisition unit and the user's mouth.

[0010] In this embodiment, the sound pressure correction unit may calculate the correction amount of the sound pressure obtained by at least one of the first sound pressure obtaining unit and the second sound pressure obtaining unit based on the difference and a correction amount corresponding mapping, function or learner representing the relationship between the difference and the correction amount of the sound pressure, and calculate the corrected sound pressure by adding the calculated correction amount to the sound pressure obtained by at least one of the first sound pressure obtaining unit and the second sound pressure obtaining unit.

[0011] In this aspect, the device may further include a determination unit for determining whether the source of the sound pressure is the user's utterance based on the ratio of the sound pressures acquired by the first sound pressure acquisition unit and the second sound pressure acquisition unit.

[0012] In this embodiment, it may be that it further comprises: an acceleration detection unit, which is arranged on the terminal body worn by the user, and detects the acceleration of the terminal body; a calculation unit, which calculates at least one of the amplitude and period of the terminal body based on the acceleration detected by the acceleration detection unit; and a correction unit, which corrects at least one of the amplitude and period of the terminal body calculated by the calculation unit based on the difference.

[0013] One solution of the present invention for achieving the above object may also be a sound analysis method, comprising the following steps:

[0014] Acquiring the sound pressure of the user's voice using a first sound pressure acquisition unit and a second sound pressure acquisition unit, each of which is disposed on equipment worn by the user and is disposed at different distances from the user's mouth when the user wears the equipment;

[0015] estimating a distance between the first sound pressure acquiring unit or the second sound pressure acquiring unit and the user's mouth based on the sound pressure acquired by the first sound pressure acquiring unit and the sound pressure acquired by the second sound pressure acquiring unit; and

[0016] A difference between a reference value of the distance between the first sound pressure acquiring unit or the second sound pressure acquiring unit and the user's mouth and the estimated distance is calculated, and the sound pressure acquired by at least one of the first sound pressure acquiring unit and the second sound pressure acquiring unit is corrected based on the difference.

[0017] One solution of the present invention for achieving the above-mentioned object may also be a program for causing a computer to execute the following processing:

[0018] Acquiring the sound pressure of the user's voice using a first sound pressure acquisition unit and a second sound pressure acquisition unit, each of which is disposed on equipment worn by the user and is disposed at different distances from the user's mouth when the user wears the equipment;

[0019] estimating a distance between the first sound pressure acquiring unit or the second sound pressure acquiring unit and the user's mouth based on the sound pressure acquired by the first sound pressure acquiring unit and the sound pressure acquired by the second sound pressure acquiring unit; and

[0020] A difference between a reference value of the distance between the first sound pressure acquiring unit or the second sound pressure acquiring unit and the user's mouth and the estimated distance is calculated, and the sound pressure acquired by at least one of the first sound pressure acquiring unit and the second sound pressure acquiring unit is corrected based on the difference.

[0021] According to the present invention, it is possible to provide a speech analysis system, a speech analysis method, and a program capable of performing speech analysis with high accuracy while suppressing a decrease in the detection accuracy of sound pressure.

[0022] The above and other objects, features and advantages of the present disclosure will be more fully understood from the detailed description given below and the accompanying drawings which are given by way of illustration only and thus should not be considered as limiting the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a block diagram showing a schematic system configuration of the sound analysis system according to the first embodiment.

[0024] Figure 2 is a diagram showing a terminal main body.

[0025] Figure 3 This is a block diagram showing a schematic system configuration of the information processing device according to the first embodiment.

[0026] Figure 4 It is a graph showing the characteristics of sound pressure.

[0027] Figure 5 This is a diagram showing an example of a distance correspondence map.

[0028] Figure 6 : is a diagram showing an example of a correction amount correspondence map.

[0029] Figure 7 This is a flowchart showing an example of the flow of the sound analysis method according to the first embodiment.

[0030] Figure 8This is a diagram showing a terminal main body according to the second embodiment.

[0031] Fig. 9 This is a block diagram showing a schematic system configuration of an information processing device according to the second embodiment.

[0032] Fig.10 This is a diagram showing a configuration in which a speech determination unit, a distance estimation unit, and a sound pressure correction unit are provided in a terminal body. DETAILED DESCRIPTION

[0033] Implementation Method 1

[0034] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Figure 1 This is a block diagram showing a schematic system configuration of a speech analysis system according to Embodiment 1. The speech analysis system 1 according to the present embodiment includes a terminal body 2 and an information processing device 3 connected to the terminal body 2 via a wireless communication line.

[0035] Wireless communication lines include Wi-Fi (registered trademark) (Wireless Fidelity), Bluetooth (registered trademark), UWB (Ultra Wideband), etc. The terminal body 2 and the information processing device 3 may also be connected to each other through a communication network such as the Internet. Alternatively, multiple terminal bodies 2 and information processing devices 3 may be connected to each other through a communication network.

[0036] For example Figure 2 As shown, the equipment worn by the user is constituted as a wearable terminal in which the terminal body 2 is suspended from the neck. A lifting rope is provided on the terminal body 2. The user can pass the neck through the lifting rope and hang the terminal body 2 from the neck to wear it.

[0037] The terminal body 2 includes first and second sound pressure acquisition units 21 and 22 for acquiring sound pressure of surrounding sounds such as user's voice, and a data transmission unit 23 for transmitting the sound pressure acquired by the first and second sound pressure acquisition units 21 and 22 to the information processing device 3 .

[0038] In the terminal body 2, the first sound pressure acquisition unit 21 and the second sound pressure acquisition unit 22 are arranged at a predetermined distance. The first and second sound pressure acquisition units 21 and 22 are a specific example of the first and second sound pressure acquisition units. The second sound pressure acquisition unit 22 is arranged at a position farther from the user's mouth than the first sound pressure acquisition unit 21 when the user hangs the terminal body 2 from the neck and wears it.

[0039] It should be noted that the first sound pressure acquisition unit 21 may also be arranged at a position farther from the user's mouth than the second sound pressure acquisition unit 22 when the user wears the terminal body 2 hanging from the neck. At least one of the first and second sound pressure acquisition units 21 and 22 may also be provided on a carrying strap or the like.

[0040] The first and second sound pressure acquisition units 21 and 22 are composed of microphones etc. for collecting voice etc. The first and second sound pressure acquisition units 21 and 22 output the acquired sound pressure to the data transmission unit 23. The data transmission unit 23 transmits the sound pressure data output from the first and second sound pressure acquisition units 21 and 22 to the information processing device 3.

[0041] The information processing device 3 has a hardware structure of a conventional computer including, for example, a processor 3a such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit), internal memories 3b such as a RAM (Random Access Memory) and a ROM (Read Only Memory), storage devices 3c such as an HDD (Hard Disk Drive) and an SDD (Solid State Drive), an input / output I / F 3d for connecting to peripheral devices such as a display, and a communication I / F 3e for communicating with devices outside the device.

[0042] The information processing device 3 can realize each function described later by, for example, the processor 3 a executing a program stored in the storage device 3 c , the internal memory 3 b , etc., while using the internal memory 3 b .

[0043] Figure 3 1 is a block diagram showing a schematic system configuration of the information processing device of this embodiment 1. The information processing device 3 includes a speech determination unit 31 for determining a speaker, a distance estimation unit 32 for estimating the distance between the first sound pressure acquisition unit 21 and the user's mouth, and a sound pressure correction unit 33 for correcting the sound pressure.

[0044] The speech determination unit 31 determines whether the source of the sound pressure output from the first and second sound pressure acquisition units 21 and 22 (hereinafter referred to as the sound pressure source) is the user wearing the terminal body 2 (hereinafter referred to as the wearing user). In other words, the speech determination unit 31 determines whether the wearing user speaks. Through this determination, the sound pressure source can be determined as the wearing user, and more accurate sound pressure correction can be performed.

[0045] like Figure 4As shown, the sound pressure has a characteristic of attenuating according to the distance from the sound pressure source. Therefore, the sound pressure ratio when the wearing user speaks and the sound pressure source is close is greater than the sound pressure ratio when other users speak and the sound pressure source is far away.

[0046] The sound pressure of the first sound pressure acquisition unit 21 when the distances between the first and second sound pressure acquisition units 21 and 22 and the sound pressure generation source are close is V 1N , the sound pressure of the second sound pressure acquisition unit 22 is set to V 2N , the distance between the first sound pressure acquisition unit 21 and the sound pressure source is set to R 1N , the distance between the second sound pressure acquisition unit 22 and the sound pressure source is set to R 2N In addition, the sound pressure of the first sound pressure acquisition unit 21 when the distance between the first and second sound pressure acquisition units 21 and 22 and the sound pressure generation source is far is set to V 1F , the sound pressure of the second sound pressure acquisition unit 22 is set to V 2F , the distance between the first sound pressure acquisition unit 21 and the sound pressure source is set to R 1F , the distance between the second sound pressure acquisition unit 22 and the sound pressure source is set to R 2F .

[0047] In this case, if Figure 4 As shown in FIG. 1 , the sound pressure ratio V when the distances between the first and second sound pressure acquisition units 21 and 22 and the sound pressure generation source are close. 1N / V 2N The sound pressure ratio V is greater than the distance between the first and second sound pressure acquisition units 21 and 22 and the sound pressure generation source. 1F / V 2F Large (V 1N / V 2N >V 1F / V 2F ).

[0048] Using such characteristics of sound pressure, the speech determination unit 31 determines whether the sound pressure generation source is the wearing user based on the ratio of the sound pressures output from the first and second sound pressure acquisition units 21 and 22 .

[0049] For example, the speech determination unit 31 calculates a first integrated value obtained by integrating the sound pressure output from the first sound pressure acquisition unit 21 during a predetermined time Δt. The speech determination unit 31 calculates a second integrated value obtained by integrating the sound pressure output from the second sound pressure acquisition unit 22 during a predetermined time Δt. The predetermined time Δt is a time when a portion of the time when the user speaks is extracted, and the time is preset for the first and second sound pressure acquisition units 21 and 22. When the speech determination unit 31 determines that the ratio of the first integrated value to the second integrated value is greater than a preset threshold value, it determines that the source of the sound pressure is the wearing user.

[0050] As described above, the speech determination unit 31 compares the ratio of the integrated values ​​of the sound pressures obtained by the first and second sound pressure obtaining units 21 and 22 with the threshold value to determine the sound pressure generation source, but the present invention is not limited thereto, and any determination method may be applied. For example, the speech determination unit 31 may compare the ratio of the average values ​​of the sound pressures obtained by the first and second sound pressure obtaining units 21 and 22 with the threshold value to determine the sound pressure generation source. Furthermore, the speech determination unit 31 may compare the difference between the integrated values ​​or average values ​​of the sound pressures obtained by the first and second sound pressure obtaining units 21 and 22 with the threshold value to determine the sound pressure generation source.

[0051] The distance estimating unit 32 estimates the distance between the first sound pressure acquiring unit 21 and the mouth of the wearer. The distance estimating unit 32 is a specific example of a distance estimating unit. Here, the sound pressure v has the property of being determined by a function (v=f(V, R)) with the volume V of the sound pressure generating source and the distance R between the sound pressure generating source and the sound pressure acquiring unit as variables. Therefore, by using two independent sound pressures (v1, v2), the distance R between the sound pressure generating source and the sound pressure acquiring unit can be uniquely determined.

[0052] Therefore, the distance estimating unit 32 estimates the distance R between the first sound pressure acquiring unit 21 and the mouth of the wearer based on the sound pressure v1 acquired by the first sound pressure acquiring unit 21 and the sound pressure v2 acquired by the second sound pressure acquiring unit 22 and a preset distance correspondence map.

[0053] Figure 5 is a diagram showing an example of distance correspondence mapping. Figure 5 As shown, the actual distance between the mouth of the wearer and the first sound pressure acquisition unit 21 is set to R, and the sound pressures v1 and v2 obtained by the first and second sound pressure acquisition units 21 and 22 at this time are associated with the distance R to create a distance correspondence map. The distance correspondence map may also be pre-set in the distance estimation unit 32.

[0054] For example, when the sound pressure v1 obtained by the first sound pressure obtaining unit 21 is 3.0 and the sound pressure v2 obtained by the second sound pressure obtaining unit 22 is 2.8, the distance estimating unit 32 refers to Figure 5 The distance correspondence map shown in the figure estimates that the distance between the first sound pressure acquisition unit 21 and the mouth of the wearer is R=4.2 cm.

[0055] The distance estimating unit 32 may also estimate the distance R between the first sound pressure acquiring unit 21 and the mouth of the user wearing the wearable device based on the sound pressures v1 and v2 acquired by the first and second sound pressure acquiring units 21 and 22 and a pre-set function. The above-mentioned function R=f(v1, v2) representing the relationship between the distance R between the user's mouth and the first sound pressure acquiring unit 21 and the sound pressures v1 and v2 acquired by the first and second sound pressure acquiring units 21 and 22 may also be set in the distance estimating unit 32.

[0056] The distance estimating unit 32 may estimate the distance R between the first sound pressure acquiring unit 21 and the mouth of the user wearing the wearer using a learner that has learned the relationship between the distance R between the user's mouth and the first sound pressure acquiring unit 21 and the sound pressures v1 and v2 acquired by the first and second sound pressure acquiring units 21 and 22 .

[0057] The learner performs machine learning by using the sound pressures v1 and v2 acquired by the first and second sound pressure acquisition units 21 and 22 as input values ​​of the learner and the distance R between the user's mouth and the first sound pressure acquisition unit 21 as output of the learner.

[0058] The learner is composed of a neural network such as RNN (Recurrent neural network). The RNN may have LSTM (Long Short Term Memory) in the middle layer. The learner may be composed of other learners such as SVM (Support Vector Machine) instead of the neural network.

[0059] The sound pressure correction unit 33 corrects at least one of the sound pressures v1 and v2 obtained by the first and second sound pressure acquisition units 21 and 22. The sound pressure correction unit 33 is a specific example of a sound pressure correction unit. For example, the sound pressure correction unit 33 calculates the difference ΔR between the reference value of the distance between the first sound pressure acquisition unit 21 and the mouth of the wearing user and the distance R estimated by the distance estimation unit 32. The reference value of the distance between the first sound pressure acquisition unit 21 and the mouth of the wearing user (hereinafter referred to as the distance reference value) is, for example, the distance between the first sound pressure acquisition unit 21 and the mouth of the wearing user measured when the terminal body 2 is suspended straight from the neck without twisting or the like using a lifting rope. The distance reference value is pre-set in the sound pressure correction unit 33.

[0060] The sound pressure correction unit 33 calculates the correction amount Δv of the sound pressure obtained by the first and second sound pressure acquisition units 21 and 22 based on the calculated difference ΔR and the correction amount correspondence map. The correspondence relationship between the difference ΔR and the correction amount Δv of the sound pressure obtained by the first and second sound pressure acquisition units 21 and 22 is obtained in advance through experiments and is set in the sound pressure correction unit 33 as a correction amount correspondence map. Figure 6 : is a diagram showing an example of a correction amount correspondence map.

[0061] The sound pressure correction unit 33 calculates the corrected sound pressures of the first and second sound pressure acquisition units 21 and 22 (hereinafter referred to as corrected sound pressures) by adding the calculated correction amount Δv to the sound pressures v1 and v2 acquired by the first and second sound pressure acquisition units 21 and 22 .

[0062] For example, when the difference ΔR is 0.5, Figure 6 As shown, the sound pressure correction unit 33 refers to the correction amount correspondence map and sets the correction amount Δv to 0.1. The sound pressure correction unit 33 adds the correction amount 0.1 to the sound pressure 3.0 acquired by the first sound pressure acquisition unit 21 to calculate the corrected sound pressure 3.1 of the first sound pressure acquisition unit 21.

[0063] The distance estimating unit 32 may also estimate the distance between the second sound pressure acquiring unit 22 and the mouth of the wearing user. In this case, the actual distance between the mouth of the wearing user and the second sound pressure acquiring unit 22 is set to R, and the sound pressures v1 and v2 acquired by the first and second sound pressure acquiring units 21 and 22 at this time are associated with the distance R to create a distance correspondence map. The distance estimating unit 32 estimates the distance R between the second sound pressure acquiring unit 22 and the mouth of the wearing user based on the distance correspondence map.

[0064] The sound pressure correction unit 33 calculates a difference ΔR between a reference value of the distance between the second sound pressure acquisition unit 22 and the mouth of the wearer and the distance R estimated by the distance estimation unit 32. The sound pressure correction unit 33 calculates a correction amount Δv of the sound pressure acquired by the first and second sound pressure acquisition units 21 and 22 based on the calculated difference ΔR and the correction amount correspondence map.

[0065] The sound pressure correction unit 33 may calculate the correction amount Δv of the sound pressure acquired by the first and second sound pressure acquisition units 21 and 22 based on the calculated difference ΔR and a function indicating the relationship between the difference ΔR and the correction amount Δv.

[0066] The sound pressure correction unit 33 may also use a learner that has learned the relationship between the difference ΔR and the correction amount Δv to calculate the correction amount Δv of the sound pressure obtained by the first and second sound pressure acquisition units 21 and 22. The difference ΔR is set as an input value of the learner, and the correction amount Δv of the sound pressure of the first and second sound pressure acquisition units 21 and 22 is set as an output of the learner, and the learner performs machine learning.

[0067] The sound pressure correction unit 33 calculates the corrected sound pressures of the first and second sound pressure acquisition units 21 and 22 by adding the calculated correction amount Δv to the sound pressures obtained by the first and second sound pressure acquisition units 21 and 22. The sound pressure correction unit 33 may also calculate the corrected sound pressures of the first or second sound pressure acquisition units 21 and 22 by adding the calculated correction amount Δv to the sound pressures obtained by the first or second sound pressure acquisition units 21 and 22.

[0068] For example, in an environment such as when the speaker is determined to be the wearer, the information processing device 3 may be configured not to include the speech determination unit 31. In this case, the sound pressure generation source is not determined, the distance estimation unit 32 estimates the distance between the first sound pressure acquisition unit 21 and the mouth of the wearer, and the sound pressure correction unit 33 calculates the corrected sound pressures of the first and second sound pressure acquisition units 21 and 22. Thus, the processing is further simplified.

[0069] Next, the sound analysis method according to the first embodiment will be described. Figure 7 This is a flowchart showing an example of the flow of the sound analysis method according to the first embodiment.

[0070] The first and second sound pressure acquisition units 21 and 22 acquire the sound pressure of the user (step S101), and output it to the data transmission unit 23. The data transmission unit 23 transmits the sound pressure output from the first and second sound pressure acquisition units 21 and 22 to the information processing device 3.

[0071] The speech determination unit 31 determines whether the sound pressure generation source is the wearing user based on the ratio of the sound pressures output from the first and second sound pressure acquisition units 21 and 22 (step S102 ).

[0072] When determining that the sound pressure generating source is not the wearing user (No in step S102 ), the speech determination unit 31 ends the present process.

[0073] On the other hand, when the speech determination unit 31 determines that the sound pressure generating source is the wearing user (step S102), the distance estimation unit 32 estimates the distance between the first sound pressure acquisition unit 21 and the mouth of the wearing user based on the sound pressure obtained by the first sound pressure acquisition unit 21 and the sound pressure and distance correspondence mapping obtained by the second sound pressure acquisition unit 22 (step S103).

[0074] The sound pressure correction unit 33 calculates the difference between the reference value of the distance between the first sound pressure acquisition unit 21 and the mouth of the wearer and the distance estimated by the distance estimation unit 32 (step S104). The sound pressure correction unit 33 calculates the correction amount of the sound pressure acquired by the first and second sound pressure acquisition units 21 and 22 based on the calculated difference and the correction amount correspondence map (step S105).

[0075] The sound pressure correction unit 33 calculates the corrected sound pressures of the first and second sound pressure acquisition units 21 and 22 by adding the calculated correction amount to the sound pressures acquired by the first and second sound pressure acquisition units 21 and 22 (step S106 ).

[0076] As described above, the sound analysis system 1 of the present embodiment 1 comprises: first and second sound pressure acquisition units 21, 22, which are respectively arranged on equipment worn by the user, and are respectively arranged at positions of different distances from the user's mouth when the user wears the equipment, and respectively acquire the sound pressure of the user's voice; a distance estimation unit 32, which estimates the distance between the first or second sound pressure acquisition unit 21, 22 and the user's mouth based on the sound pressure obtained by the first sound pressure acquisition unit 21 and the sound pressure obtained by the second sound pressure acquisition unit 22; and a sound pressure correction unit 33, which calculates the difference between a reference value of the distance between the first or second sound pressure acquisition unit 21, 22 and the user's mouth and the distance estimated by the distance estimation unit 32, and corrects the sound pressure obtained by at least one of the first and second sound pressure acquisition units 21, 22 based on the difference.

[0077] Thus, even when the distance between the user's mouth and the first and second sound pressure acquisition units 21 and 22 changes, the sound pressure is appropriately corrected according to the changed distance. Therefore, it is possible to suppress a decrease in the sound pressure detection accuracy and perform speech analysis with high accuracy.

[0078] Implementation Method 2

[0079] In this embodiment 2, if Figure 8 As shown, in addition to the first and second sound pressure acquisition units 21 and 22, an acceleration sensor 24 is also provided in the terminal body 20. The acceleration sensor 24 detects the acceleration of the terminal body 20. The amplitude and period of the terminal body 20 are calculated based on the acceleration detected by the acceleration sensor 24, and the action (nodding, etc.) of the wearing user is estimated. At this time, according to the principle of the pendulum, even if the action is the same, if the length of the lifting rope changes, the amplitude and period of the terminal body 20 will also change. Therefore, it is preferred to correct the amplitude and period of the terminal body 20 according to the length of the lifting rope.

[0080] The speech analysis system of the second embodiment corrects at least one of the amplitude and the period of the terminal body 20 based on the difference ΔR that changes according to the length of the lifting cord.

[0081] Fig. 9 1 is a block diagram showing a schematic system configuration of an information processing device according to Embodiment 2. The information processing device 30 according to Embodiment 2 includes an amplitude calculation unit 34, an amplitude correction unit 35, a period calculation unit 36, and a period correction unit 37 in addition to the above-mentioned speech determination unit 31, distance estimation unit 32, and sound pressure correction unit 33.

[0082] The amplitude calculation unit 34 calculates the amplitude of the terminal body 20 based on the acceleration detected by the acceleration sensor 24. The amplitude calculation unit 34 is a specific example of a calculation unit. The amplitude correction unit 35 corrects the amplitude calculated by the amplitude calculation unit 34. The amplitude correction unit 35 is a specific example of a correction unit.

[0083] For example, the amplitude correction unit 35 calculates the correction amount of the amplitude calculated by the amplitude calculation unit 34 based on the difference ΔR and the correction amount correspondence map. The correspondence relationship between the difference ΔR and the correction amount of the amplitude calculated by the amplitude calculation unit 34 is obtained in advance through experiments and is set in the amplitude correction unit 35 as a correction amount correspondence map. It should be noted that the amplitude correction unit 35 may also calculate the correction amount of the amplitude using a function or a learner that represents the relationship between the difference ΔR and the correction amount of the amplitude calculated by the amplitude calculation unit 34. The amplitude correction unit 35 calculates the corrected amplitude by adding the calculated correction amount to the amplitude calculated by the amplitude calculation unit 34.

[0084] Similarly, the period calculation unit 36 ​​calculates the period of the terminal body 20 based on the acceleration detected by the acceleration sensor 24. The period calculation unit 36 ​​is a specific example of a calculation unit. The period correction unit 37 corrects the period calculated by the period calculation unit 36. The period correction unit 37 is a specific example of a correction unit.

[0085] For example, the period correction unit 37 calculates the correction amount of the period calculated by the period calculation unit 36 ​​based on the difference ΔR and the correction amount correspondence map. The correspondence relationship between the difference ΔR and the correction amount of the period calculated by the period calculation unit 36 ​​is obtained in advance through experiments and is set in the period correction unit 37 as a correction amount correspondence map. It should be noted that the period correction unit 37 may also calculate the correction amount of the period using a function or a learner that represents the relationship between the difference ΔR and the correction amount of the period calculated by the period calculation unit 36. The period correction unit 37 calculates the corrected period by adding the above-calculated correction amount to the period calculated by the period calculation unit 36.

[0086] Furthermore, sensors such as a heart rate sensor and a step sensor other than the acceleration sensor 24 may also be provided in the terminal body 20. In this case, too, when the value obtained by the sensor changes according to the distance from the mouth of the wearing user, it can be corrected by the same method as described above.

[0087] In the second embodiment, the same reference numerals are used to denote the same components as those in the first embodiment, and detailed description thereof will be omitted.

[0088] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other ways, and various omissions, substitutions, and changes can be made without departing from the scope of the subject matter of the invention. These embodiments and their variations are included in the scope and subject matter of the invention, and are included in the invention described in the claims and their equivalents.

[0089] For example, in the above-mentioned embodiment, at least one of the speech determination unit 31, the distance estimation unit 32, the sound pressure correction unit 33, the amplitude calculation unit 34, the amplitude correction unit 35, the period calculation unit 36 ​​and the period correction unit 37 may be provided in the terminal body 2.

[0090] Fig.10 2 is a diagram showing a configuration in which the speech determination unit, the distance estimation unit, and the sound pressure correction unit are provided in the terminal body. In this case, the processing of the information processing device 3 is not required, so the terminal body 40 may not have the data transmission unit 23. Therefore, the configuration of the sound analysis system can be further simplified.

[0091] In addition, in the above-mentioned embodiment, the terminal body 2 is configured as a wearable terminal that can be hung from the neck by a rope, but is not limited to this. The terminal body 2 can also be configured as a wearable terminal incorporated into a necklace, glasses (including sunglasses, etc.), headphones, headwear, watches, bracelets, clothing, etc. It should be noted that no matter which structure is used, as in the above-mentioned embodiments 1 and 2, the first and second sound pressure acquisition units 21 and 2 are respectively arranged at different distances from the user's mouth when the user wears the wearable terminal.

[0092] The present invention can also be used, for example, Figure 6 The illustrated processing is realized by causing the processor 3a to execute a computer program.

[0093] The program can be stored and provided to the computer using various types of non-transitory computer readable media. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer readable media include magnetic recording media (e.g., floppy disks, magnetic tapes, hard disk drives), optical magnetic recording media (e.g., optical magnetic disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (random access memory)).

[0094] The program can also be supplied to the computer by various types of transitory computer readable media. Examples of transitory computer readable media include electrical signals, optical signals, and electromagnetic waves. Transitory computer readable media can supply the program to the computer via wired communication paths such as wires and optical fibers or wireless communication paths.

[0095] Each component constituting the information processing device 3 of each embodiment described above can be realized not only by a program but also partially or entirely by dedicated hardware such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array).

[0096] From the disclosure thus described, it is obvious that the embodiments of the disclosure may be modified in a variety of ways. Such variations should not be regarded as departing from the spirit and scope of the disclosure, and all such modifications obvious to those skilled in the art are intended to be included within the scope of the appended claims.

Claims

1. A sound analysis system comprising: The first sound pressure acquisition unit and the second sound pressure acquisition unit are respectively arranged on the equipment worn by the user, and are respectively arranged at different distances from the user's mouth when the user wears the equipment, and respectively acquire the sound pressure of the user's voice; a distance estimating unit that estimates the distance between the first sound pressure acquiring unit and the user's mouth based on the sound pressure acquired by the first sound pressure acquiring unit and the sound pressure acquired by the second sound pressure acquiring unit, or estimates the distance between the second sound pressure acquiring unit and the user's mouth based on the sound pressure acquired by the first sound pressure acquiring unit and the sound pressure acquired by the second sound pressure acquiring unit; a sound pressure correction unit that calculates a difference between a reference value of the distance between the first sound pressure acquisition unit or the second sound pressure acquisition unit and the user's mouth and the distance estimated by the distance estimation unit, and corrects the sound pressure acquired by at least one of the first sound pressure acquisition unit and the second sound pressure acquisition unit based on the difference; and The speech determination unit determines whether the source of the sound pressure is the user based on the ratio of the sound pressures acquired by the first sound pressure acquisition unit and the second sound pressure acquisition unit.

2. The sound analysis system according to claim 1, The distance estimation unit estimates the distance between the first sound pressure acquisition unit or the second sound pressure acquisition unit and the user's mouth based on the sound pressure acquired by the first sound pressure acquisition unit and the second sound pressure acquisition unit and a distance correspondence mapping, function or learner representing the relationship between the sound pressure acquired by the first sound pressure acquisition unit and the second sound pressure acquisition unit and the distance between the first sound pressure acquisition unit or the second sound pressure acquisition unit and the user's mouth.

3. The sound analysis system according to claim 1 or 2, The sound pressure correction unit calculates the correction amount of the sound pressure obtained by at least one of the first sound pressure acquisition unit and the second sound pressure acquisition unit based on the difference and a correction amount corresponding mapping, function or learner representing the relationship between the difference and the correction amount of the sound pressure, and calculates the corrected sound pressure by adding the calculated correction amount to the sound pressure obtained by at least one of the first sound pressure acquisition unit and the second sound pressure acquisition unit.

4. The sound analysis system according to claim 1 or 2, further comprising: An acceleration detection unit, disposed on a terminal body worn by the user, and detecting acceleration of the terminal body; a calculation unit that calculates at least one of an amplitude and a period of the terminal body based on the acceleration detected by the acceleration detection unit; and The correction unit corrects at least one of the amplitude and the period of the terminal body calculated by the calculation unit based on the difference.

5. A sound analysis method comprising the following steps: Acquiring the sound pressure of the user's voice using a first sound pressure acquisition unit and a second sound pressure acquisition unit, each of which is disposed on equipment worn by the user and is disposed at different distances from the user's mouth when the user wears the equipment; estimating the distance between the first sound pressure acquiring unit and the user's mouth based on the sound pressure acquired by the first sound pressure acquiring unit and the sound pressure acquired by the second sound pressure acquiring unit, or estimating the distance between the second sound pressure acquiring unit and the user's mouth based on the sound pressure acquired by the first sound pressure acquiring unit and the sound pressure acquired by the second sound pressure acquiring unit; calculating a difference between a reference value of the distance between the first sound pressure acquiring unit or the second sound pressure acquiring unit and the user's mouth and the estimated distance, and correcting the sound pressure acquired by at least one of the first sound pressure acquiring unit and the second sound pressure acquiring unit based on the difference; and Whether the source of the sound pressure is the user is determined based on a ratio of the sound pressures acquired by the first sound pressure acquiring unit and the second sound pressure acquiring unit.

6. A computer-readable medium storing a program for causing a computer to execute the following processing: Acquiring the sound pressure of the user's voice using a first sound pressure acquisition unit and a second sound pressure acquisition unit, each of which is disposed on equipment worn by the user and is disposed at different distances from the user's mouth when the user wears the equipment; estimating the distance between the first sound pressure acquiring unit and the user's mouth based on the sound pressure acquired by the first sound pressure acquiring unit and the sound pressure acquired by the second sound pressure acquiring unit, or estimating the distance between the second sound pressure acquiring unit and the user's mouth based on the sound pressure acquired by the first sound pressure acquiring unit and the sound pressure acquired by the second sound pressure acquiring unit; calculating a difference between a reference value of the distance between the first sound pressure acquiring unit or the second sound pressure acquiring unit and the user's mouth and the estimated distance, and correcting the sound pressure acquired by at least one of the first sound pressure acquiring unit and the second sound pressure acquiring unit based on the difference; and Whether the source of the sound pressure is the user is determined based on a ratio of the sound pressures acquired by the first sound pressure acquiring unit and the second sound pressure acquiring unit.

Images