JP2013142795A - Conversation protection system and conversation protection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent that the content of conversation is caught by a third party by reproducing sound, and to reduce discomfort of the third party to the reproduced sound.SOLUTION: A conversation protection system is constituted of: a microphone for collecting voice for performing conversation; a storage part which stores attention sound effect data showing time axis waveform which attenuates after a sound pressure level starts in several seconds and base sound effect data showing time axis waveform whose sound pressure level gradually changes in comparison with an attention sound effect; a speaker for reproducing the attention sound effect and the base sound effect toward a third party except a party concerned; and a control part which reproduces the base sound effect while at least the sound pressure level of conversation voice collected by the microphone exceeds a first threshold, and reproduces the attention sound effect whenever the sound pressure level of the conversation voice exceeds a second threshold.

Description

  The present invention relates to a conversation protection system and a conversation protection method for preventing the contents of a conversation from being heard by a third party other than the conversation party.

  2. Description of the Related Art Conventionally, various systems are used in banks, hospitals, and the like to prevent a conversational voice from leaking and being heard by a third party. For example, a conversation between a bank employee and a customer in a bank or a conversation between a receptionist, a doctor, a pharmacist, and a patient in a hospital may contain personal information that a third party does not want to hear. For this reason, by playing another sound toward the third party, it is difficult for the third party to hear the conversation voice.

  For example, in Patent Document 1, the content of a conversation is made by flowing a BGM (Background Music) after a screen having a sound absorbing effect is set as a partition between a plurality of groups that have a conversation in a public place such as a station or an airport. A technique for protecting the image is disclosed. Patent Document 2 discloses a technique for collecting conversational speech performed in a neighboring room and adjusting the volume of the BGM in accordance with the volume of the conversational speech with respect to BGM reproduction.

  However, the human ear has a selective listening ability to selectively listen to a specific sound by the so-called cocktail party effect. For this reason, in order to make it more difficult to listen to the conversational sound, a masking sound generated based on the conversational sound may be used instead of the BGM unrelated to the conversation.

  For example, Patent Document 3 discloses a technique of generating a sound having a phase opposite to the frequency spectrum of conversational speech and reproducing it as a masking sound only during conversation. Patent Document 4 discloses a technique that uses hearing loss generated based on an envelope and a fine structure extracted from a frequency spectrum of a conversational sound for masking sound. The hearing-proof sound is a type of masking sound that can prevent the conversation content from being listened to and can break the phonological property of the conversation voice by playing it over the conversation voice.

Special table 2011-528445 gazette JP 2007-256606 A JP 2010-19935 A Japanese Patent No. 4761506

  However, according to the above prior art, a third party may feel uncomfortable or uncomfortable with respect to BGM or masking sound for protecting the content of conversation. For example, when music or the like is reproduced as BGM, taking into account the cocktail party effect, if the volume of the BGM is increased so that the conversational voice is not heard, the loud sound may be uncomfortable.

  In addition, when reproducing the masking sound, there is a case where the sound having the artificially generated frequency characteristic is uncomfortable. Since the masking sound is generated according to the characteristics of the conversational voice, it is possible to obtain an effect that makes it difficult to hear the conversational voice at a lower volume than when using a sound unrelated to the conversational voice such as BGM. Can do. However, a third party who hears an unfamiliar masking sound that is not experienced in daily life may feel uncomfortable even if the volume is not high. In addition, the user may concentrate on listening to an unfamiliar masking sound while feeling uncomfortable, which may increase the uncomfortable feeling.

  The present invention has been made to solve the above-described problems caused by the prior art, and while reproducing the sound for protecting the conversation contents so that the conversation voice cannot be heard by a third party, It is an object of the present invention to provide a conversation protection system and a conversation protection method in which a third party does not feel discomfort or discomfort.

  In order to solve the above-mentioned problems and achieve the object, the present invention is a conversation protection system, which is a microphone for collecting speech to be spoken and attenuating after a sound pressure level rises in a few seconds A storage unit for storing an attention effect sound indicating a time axis waveform to be performed and a base effect sound indicating a time axis waveform in which the sound pressure level gradually changes compared to the attention effect sound, and a conversation party excluding the conversation party performing the conversation A sound pressure level of a conversational sound collected by at least a speaker for reproducing one or both of the attention sound effect and the base sound effect and at least the microphone is set to a first threshold. While the value exceeds the value, control for reproducing the base sound effect by the speaker and the attention effect each time the sound pressure level of the conversational sound exceeds a second threshold value. Characterized in that it comprises a control unit for performing either one or both control of the control to play sound.

  Also, in the present invention according to the above-mentioned invention, the control unit may select one of the base sound effect and the attention sound effect by the speaker when the sound pressure level of the conversational sound falls below a predetermined threshold value. Alternatively, both reproductions are stopped.

  Further, according to the present invention, in the above invention, each time the attention sound effect is reproduced, the control unit counts the number of sounds that can be heard within a predetermined time, the timing at which the sound is heard, the tone color of the sound, and the pitch of the sound. The reproduction of the attention sound effect is controlled so that at least one of them changes.

  Further, the present invention is characterized in that, in the above invention, the control unit generates the attention sound effect using a sine wave.

  The present invention further includes a masking sound generation unit that generates a masking sound that masks the sound and makes it difficult to hear based on a frequency characteristic of the sound collected by the microphone. The masking sound generated by the masking sound generator is reproduced.

  Further, the present invention is the above invention, wherein the masking sound generator extracts a spectral envelope and a spectral fine structure from the sound collected by the microphone, and raises and lowers the spectral envelope around an axis extending in a frequency direction. An inversion axis is set to replace the spectrum envelope, and the spectrum envelope is inverted about the inversion axis to deform the spectrum envelope to generate a deformed spectrum envelope, and the deformed spectrum envelope and the spectrum fineness are generated. A hearing loss sound having a synthesized structure is generated and used as the masking sound.

  Further, the present invention is the above invention, wherein the storage unit stores a plurality of the attention sound effects, and the control unit reproduces the attention sound effects randomly selected from the storage unit. Features.

  Further, the present invention is the above invention, wherein the storage unit stores a plurality of attention sound effect combinations set based on the tone color and pitch of each attention sound effect as a setting table, The control unit plays the attention sound effect based on a combination randomly selected from the setting table.

  Further, the present invention is characterized in that, in the above-mentioned invention, the control unit randomly changes a volume when reproducing each attention sound effect.

  In the present invention, the attention sound effect is a sound of a musical instrument.

  The present invention is also a conversation protection method, comprising: a sound collecting step for collecting conversational sound; and a sound pressure level of at least the conversation sound collected in the sound collecting step has a first threshold value. The sound pressure level of the conversational sound collected in the voice sound collection step and the base sound effect reproduction step for reproducing the base sound effect showing the time axis waveform in which the sound pressure level changes slowly, One or both of the attention effect sound reproducing steps for reproducing the attention effect sound indicating the time axis waveform that decays after the sound pressure level rises within a few seconds when the threshold value of 2 is exceeded And a sound effect reproduction step including.

  Further, the present invention provides the above-described invention, wherein, when the sound pressure level of the conversational sound collected in the voice sound collecting step falls below a predetermined threshold value, one of the base sound effect and the attention sound effect. It further includes a sound effect stop step for stopping one or both reproductions.

  According to the present invention, at least the base sound effect is reproduced while the sound pressure of the conversation voice to be protected exceeds a predetermined threshold, and the attention sound effect is generated at random timing according to the volume of the conversation voice. It is possible to give a random impression to the listener by playing. In addition, attention sound effects, which are short sounds that decay within a few seconds, are easy to draw the attention of third parties, so that third parties who are used to the sound can listen to conversational voices by selectively listening to the cocktail party effect. Can prevent and enhance the effect of conversation protection.

  Further, according to the present invention, when there is no conversation, the bass sound effect and the attention sound effect can be stopped, which is suitable for use in a quiet environment.

  Further, according to the present invention, at least one of the number of sounds that can be heard within a predetermined time by one reproduction of the attention sound effect, the timing at which the sound is heard, the tone color of the sound, and the pitch of the sound is changed. As a result, a random impression can be given to a third party to prevent them from getting used to the sound.

  Further, according to the present invention, attention sound effects can be generated using sine waves, so attention sound effects that give a random impression to the listener can be freely generated and used.

  Further, according to the present invention, it is possible to protect the conversation by reproducing the masking sound generated based on the frequency characteristic of the conversation voice. In addition, even if the masking sound is an uncomfortable sound that cannot be heard in daily life, the third party's attention can be directed to the base sound effect and the attention sound effect, which can reduce the uncomfortable feeling of the masking sound. it can.

  Further, according to the present invention, it is possible to protect the conversation more effectively by reproducing the hearing-aid sound for breaking the phoneme of the conversation voice. In addition, the sense of incongruity of the hearing-proof sound can be reduced by the base effect sound and the attention effect sound.

  Further, according to the present invention, various attention sound effect data having different timbres, pitches, etc. are stored in the storage unit, and a random impression is given to the listener by playing back a randomly selected sound from among them. Thus, it is possible to maintain the effect of protecting the conversation and reducing the uncomfortable feeling of the masking sound such as the hearing-aid sound.

  In addition, according to the present invention, a combination of attention sound effects that does not give a sense of incongruity even when continuously played back is set as a setting table based on the tone color and the attention sound effects are played back based on this setting. In addition to the effect of protecting the conversation and the effect of reducing the uncomfortable feeling of the masking sound, the uncomfortable feeling of the attention sound can also be reduced to give a pleasant impression.

  Further, according to the present invention, in addition to changing the tone color of the attention sound effect to be reproduced, the volume at the time of reproducing the attention sound effect can also be changed. Third parties can get used to the sound and suppress the effect of the cocktail party.

FIG. 1 is a diagram for explaining an example of use of a conversation protection system according to the present invention. FIG. 2 is a diagram for explaining an installation example of the conversation protection system according to the present embodiment. FIG. 3 is a block diagram illustrating a schematic functional configuration of the conversation protection system according to the present embodiment. FIG. 4 is an explanatory diagram illustrating an analysis example of conversational speech according to the present embodiment. FIG. 5 is a diagram for explaining the reproduction timing of the bass sound effect and the attention sound effect according to the present embodiment. FIG. 6 is a diagram illustrating an example of the amplitude waveform of the attention sound effect according to the present embodiment. FIG. 7 is a diagram illustrating an example of an attention sound effect stored in the storage unit according to the present embodiment. FIG. 8 is a diagram illustrating an example of a setting table in which a combination of attention sound effects stored in the storage unit according to the present embodiment is set. FIG. 9 is a diagram for explaining a method of reproducing a plurality of attention sound effects according to the present embodiment. FIG. 10 is a flowchart for explaining a bass sound effect reproducing method according to the present embodiment. FIG. 11 is a flowchart for explaining a method for reproducing an attention sound effect according to the present embodiment. FIG. 12 is a diagram for explaining a method for generating an attention sound effect according to this embodiment.

  Exemplary embodiments of a conversation protection system and a conversation protection method according to the present invention will be explained below in detail with reference to the accompanying drawings. The conversation protection system protects personal information and privacy contained in conversations by preventing third parties from listening to conversations conducted at financial institutions such as banks and medical institutions such as hospitals and pharmacies. It is a system used for this purpose.

  FIG. 1 is a diagram illustrating an example of use of a conversation protection system. In this example, the voices of the parties 1 and 2 having a conversation in the booth are prevented from being heard by the third party 3. For example, a person sitting in a waiting seat 52 outside a partition 51 provided for partitioning a booth or a person in an adjacent booth corresponds to the third party 3.

  The conversation protection system includes an audio processing device 10, a microphone (hereinafter referred to as “microphone”) 20, and a speaker 30. The microphone 20 is installed, for example, on a table 50 on which the conversation parties 1 and 2 sit in a booth, and is used to collect conversational voice to be protected. The sound processing device 10 has a function of generating a masking sound based on the sound collected by the microphone 20 and reproducing the masking sound and a sound effect described later by the speaker 30. The speaker 30 is installed so as to reproduce sound toward the third party 3 outside the booth.

  FIG. 2 is a diagram for explaining an installation example of the conversation protection system as viewed from above. As described above, the conversation protection system may include the output sound operation unit 40 for controlling the sound output from the speaker 30 in addition to the sound processing device 10, the microphone 20, and the speaker 30. By this output sound operation unit 40, for example, playback start and playback stop control of sound output from the speaker 30 and volume control during playback can be performed. In FIG. 2, only the system for protecting the conversation of one booth is shown for the sake of simplicity, but when there are a plurality of booths, the conversation voice of each booth is protected. Audio processing apparatus 10, microphone 20, speaker 30, and output sound operation unit 40 are installed.

  FIG. 3 is a block diagram showing an outline of a functional configuration of the conversation protection system. Details of the speech processing apparatus 10 will be described with reference to FIG. The speech processing apparatus 10 includes an input sound analysis unit 11 that receives input of conversational sound collected by a microphone 20, a hearing-proof sound generation unit (masking sound generation unit) 12 that generates a hearing-aid sound based on the conversational sound, A sound effect control unit 13 that controls a sound effect to be reproduced separately from the listening sound, a storage unit 14 that stores sound data used as the sound effect, settings for controlling the reproduction of the sound effect, etc. And an output sound control unit 15 that performs control to reproduce the sound effect on the speaker 30 according to the conversational sound. The input sound analysis unit 11, the hearing loss generation unit 12, the sound effect control unit 13, and the output sound control unit 15 are configured by a DSP (Digital Signal Processor). The voice processing apparatus 10 may be configured by dedicated hardware including a DSP, or may be configured using a computer apparatus.

  In FIG. 3, only the components necessary for explaining the conversation protection system are shown, but the audio processing apparatus 10 also processes, for example, an input signal from the microphone 20 and an output signal to the speaker 30. An A / D (D / A) converter and an amplifier. There may also be a communication interface for performing wired or wireless communication with an external device.

  The input sound analysis unit 11 has a function of analyzing the frequency characteristics and volume of conversational speech that is input from the microphone 20 and digitized by the A / D converter and that is to be protected. For example, when the conversational voice shows the amplitude waveform shown in the upper part of FIG. 4, the amplitude waveform is analyzed in real time to generate the sound pressure waveform shown in the lower part of the figure. The sound pressure level that forms the sound pressure waveform is used to control the process of generating the hearing loss by the hearing loss generator 12 and the reproduction timing of the sound effect generated by the sound controller 13.

  The hearing-proof sound generation unit 12 has a function of generating a hearing-proof sound that can break the phonological property of the conversation voice by playing it from the speaker 30 in accordance with the conversation voice. Hearing loss is a type of masking sound that is generated based on features related to an envelope and a fine structure extracted from a spectrum indicating frequency characteristics of conversational speech. Specifically, the spectral envelope and spectral fine structure are extracted from the speech spectrum obtained from the conversational speech, an inversion axis extending in the frequency direction for switching the spectral envelope up and down is set, and the spectrum is centered on the inversion axis. By reversing the envelope, a deformed spectrum envelope is generated by deforming the spectrum envelope. Further, a deformed spectrum obtained by synthesizing the deformed spectrum envelope and the spectrum fine structure is generated, and this is used as hearing loss. In addition, since a hearing-proof sound can be produced | generated by the prior art disclosed by the patent 4761506, detailed description is abbreviate | omitted.

  The sound effect control unit 13 has a function of generating a sound effect reproduced from the speaker 30 in accordance with a conversational sound by controlling a combination of sound effect data and a reproduction volume. Here, the sound effect referred to in the present embodiment is a sound that is reproduced in order to obtain an effect that makes it difficult for the third party 3 to hear the conversational sound and an effect that reduces the sense of incongruity with the hearing-aid sound. While the hearing-proof sound is a sound having a frequency characteristic based on the characteristics of the conversational sound, the sound effect has a frequency characteristic unrelated to the conversational sound. Further, while the hearing-proof sound is a sound generated based on the conversational sound, a sound such as a song or a musical instrument prepared in advance can be used as the sound effect. Hearing loss has artificially manipulated frequency characteristics, so it may feel uncomfortable when you hear it, but there is no such discomfort from sound effects, and depending on the type of sound, it gives a pleasant impression. receive.

  As sound effects, two types of sounds are used: bass sound effects and attention sound effects. The base sound effect is a sound that is reproduced continuously without interruption during the conversational voice, and the attention sound effect is a sound that is reproduced every time the sound pressure of the conversational voice exceeds a predetermined threshold. The base sound effect data 14b used as the base sound effect and the attention sound effect data 14a used as the attention sound effect are stored in the storage unit 14 including a general memory such as a flash memory. For example, a PCM (Pulse Code Modulation) sound source is used as the data format of the sound effect data. Also, a compressed sound source such as MP3 may be used. The storage unit 14 may use another storage device such as a hard disk as long as it can store a plurality of sound effect data.

  As the bass sound effect, it is preferable to use a song that receives a quiet impression. Specifically, a sound source with a gradual change in sound pressure level in the time axis direction is used. For example, a plurality of music boxes are stored in the storage unit 14 in advance as the base sound effect data 14b. The base sound effect is a sound corresponding to BGM used in a conventional apparatus.

  The base sound effect is several seconds to several tens of seconds long and does not include sounds that would make the third party 3 surprised or uncomfortable when played, even if the same song or another song is played continuously The content is not particularly limited as long as the end of the song is connected to the beginning of the next song without a sense of incongruity and no discomfort is felt even when the volume is changed. For example, it may be a song by a musical instrument such as a percussion instrument or a stringed instrument, or may be a sound using a river murmur, a sound of a wave, or a song of a bird.

  The user can select a sound to be reproduced from a plurality of bass sound effect data 14 b stored in the storage unit 14. When the bass sound effect data 14b to be reproduced is selected in advance, the selected content is stored in the storage unit 14 as a part of the sound effect output condition 14c. The sound effect control unit 13 selects the base sound effect data 14b with reference to the setting included in the sound effect output condition 14c. Then, the selected bass sound effect is reproduced from the speaker 30 by the output sound control unit 15. Note that the output sound operation unit 40 may include an operation unit for selecting the bass sound effect data 14b, and the user may select the base sound effect data 14b by operating the operation unit. Moreover, the aspect which the base sound effect data 14b may be selected based on the signal received from external devices, such as a remote control, with the communication function with which the audio | voice processing apparatus 10 is provided may be sufficient.

  As the attention sound, it is preferable to use a sound that is not surprising to the listener and that is easy to draw attention. For example, a short sound is used in which the sound pressure level in the time axis direction rises within a few seconds and then decays. Specifically, a sound of hitting a percussion instrument such as an iron koto or a xylophone, a sound of playing a stringed instrument such as a guitar or a harp, or a sound of a bell or bell is used as an attention effect sound. Sounds of various pitches with various pitches are stored in advance in the storage unit 14 as attention sound effect data 14a.

  Attention sound effects are short sounds that are not surprised or uncomfortable by a third party when played, but do not feel uncomfortable when played over the base sound effects and are buried in the base sound effects As long as it is a sound that can be easily heard, the conditions such as tone color and pitch are not particularly limited. For example, it may be a sound of an folk instrument such as kalimba, a sound generated using a sine wave, a bird's singing sound, an insect's crying, or the like. Further, data stored in advance in the storage unit 14 may be used, or a mode in which sound is generated by combining a plurality of sine waves to which a window function is applied may be used.

  The attention sound effect data 14 a is randomly selected by the sound effect control unit 13 from the data stored in the storage unit 14. The attention sound effect data 14a is reproduced by being superimposed on the base sound effect data 14b at a timing when the volume of the conversational sound exceeds a predetermined threshold value.

  The storage unit 14 stores sound effect output conditions 14c in addition to the base sound effect data 14b and the attention sound effect data 14a. The sound effect output condition 14c includes setting information related to the base sound effect data 14b selected by the user of the conversation protection system, the reproduction condition of the base sound effect data 14b, the selection condition and the reproduction condition of the attention sound effect data 14a, and the like. included.

  The output sound control unit 15 includes a D / A converter and an amplifier, and has a function of reproducing the hearing-aid sound generated by the hearing-aid sound generation unit 12 from the speaker 30. The hearing-proof sound is reproduced so as to destroy the phoneme of the voice over the conversation voice so that the contents of the conversation cannot be heard. Further, the output sound control unit 15 has a function of reproducing the sound effect selected by the sound effect control unit 13 from the speaker 30 at a volume set by the sound effect control unit 13.

  The playback timing of the sound effect is controlled based on the sound pressure level of the conversational voice generated by the input sound analysis unit 11 as shown in FIG. The threshold value B for the base sound effect and the threshold value A for the attention sound effect used for controlling the reproduction timing are included in the sound effect output condition 14c stored in the storage unit 14.

  FIG. 5 is a diagram for explaining the reproduction timing of the sound effect. For example, as shown in the lower part of the sound pressure waveform in FIG. 5, when the sound pressure level of the conversational voice exceeds the threshold value B, the reproduction of the base effect sound is started. When the sound pressure level falls below the threshold value B, the reproduction of the bass sound effect is stopped. The base sound effect is selected based on the setting condition included in the sound effect output condition 14c, and is reproduced from the speaker 30 by the output sound control unit 15 at a predetermined volume level.

  The base sound effect is reproduced only while the sound pressure level of the conversation voice exceeds the threshold B, and is controlled to stop when there is no conversation. For this reason, even if the conversation protection system is installed in a quiet environment, the base sound effect is played at a preset volume only when it is necessary to protect the conversation. It is possible to keep the state.

  The method for controlling the reproduction of the bass sound effect is not limited to the mode in which the reproduction of the base sound effect is stopped every time the sound pressure level of the conversational sound falls below the threshold value B. You may make it stop reproduction | regeneration of a bass sound effect only when the state in which it fell below continued for the preset predetermined time. That is, when the conversation is only interrupted for a short time, it may be controlled to continue playing the base sound effect. When it is desired to always reproduce the bass sound effect and use it as the BGM, control may be performed so that the bass sound effect continues to be played unless the playback sound operation unit 40 is operated to stop the playback. Also, instead of controlling the start and stop of the playback of the bass sound effect, the playback volume shown in FIG. 5 is controlled by constantly playing the base sound effect and controlling the playback volume according to the sound pressure level of the conversational voice. You may realize the situation. The reproduction control of these bass sound effects can be realized by changing the setting of the sound effect output condition 14c.

  Furthermore, regarding the threshold value for the base sound effect, the threshold value for controlling the reproduction of the base sound effect so that it can be heard by the third party 3 and the reproduction of the base sound effect so that it cannot be heard by the third person 3 The threshold value for control may be set to a different value. That is, for example, the threshold value for determining the start of playback of the bass sound effect and the threshold value for determining the stop of playback may be different set values.

  Further, as shown in the upper part of the sound pressure waveform in FIG. 5, after the sound level of the conversational sound exceeds the threshold value B and the reproduction of the base effect sound is started, the sound pressure level is further increased to the threshold value. When A is exceeded, an attention sound effect is played. The attention sound effect is randomly selected by the sound effect control unit 13 from a plurality of sound data stored in the storage unit 14. Also, the sound volume level at the time of reproduction is randomly set by the sound effect control unit 13. As for the volume level at the time of reproduction, only the range of the volume level is set in advance, and is set at random within this range.

  Attention sound effects are played each time the sound pressure of the conversation voice exceeds the threshold A, but since the sound pressure of the conversation voice changes irregularly, the attention sound effects are at random timing as shown in FIG. Will be played. In FIG. 5, the attention sound effect is shown in two stages. Thus, depending on the timing at which the sound pressure of the conversational sound exceeds the threshold value A, the reproduction of the previously played attention sound effect is completed. Before the next attention sound effect may be played.

  FIG. 5 shows a case where both the base sound effect and the attention sound effect are reproduced. However, the present embodiment is not limited to this, and the base value can be set by setting the threshold values A and B. It is also possible to control to reproduce only one of the sound effect and the attention sound effect.

  As an attention sound effect, a short sound such as a bell sound that can easily attract the listener's attention is used. For this reason, an attention sound effect is reproduced at a timing when the volume of the conversation sound exceeds a predetermined threshold A, and the third party's attention is directed to the attention effect sound, thereby making it difficult to hear the conversation sound. .

  In addition, although not shown in FIG. 5, the hearing protection sound generated by the hearing protection sound generation unit 12 is also reproduced as a masking sound in accordance with the conversational sound. For example, as in the case of the base sound effect shown in FIG. 5, the hearing-proof sound is reproduced in accordance with the conversation sound while the threshold value B is exceeded. A hearing-proof sound may be a sound that makes you feel uncomfortable. However, since the attention effect sound is reproduced at random timing when the conversational sound exceeds the threshold A while the hearing protection sound is reproduced, the attention of the third party is attracted to the attention effect sound. As a result, it is possible to obtain an effect of reducing a sense of incongruity with the hearing loss sound.

  Further, the attention sound effect is reproduced while changing the tone color, pitch, reproduction volume and timing, so that it does not receive a monotonous impression as in the case where a similar sound is reproduced repeatedly. That is, the random impression of the attention sound effect prevents a third party from getting used to the attention sound effect. For this reason, it is possible to keep the attention of a third party and maintain the effect of protecting the conversational sound and the effect of reducing the uncomfortable feeling of the hearing loss sound.

  Since the attention sound is a short sound that is played only when the sound pressure of the conversation voice exceeds the threshold A, the attention sound is played until the next attention sound is played after the end of the reproduction of the attention sound. There may be a gap time between the two. When only the attention effect sound is played back without playing the base sound effect, the impression of the hearing aid sound becomes strong in the gap between the attention effect sounds, and there is a possibility that the hearing aid sound may feel uncomfortable. For this reason, a threshold value B having a sound pressure level lower than the threshold value A is set, the base sound effect is reproduced while the threshold value B is exceeded, and the base sound effect is generated in the gap between the two attention sound effects. Playback is controlled so that can be heard. As a result, it is possible to avoid an increase in the impression of the hearing loss sound in the gap between the attention sound effects.

  The bass sound effect is a sound with less volume fluctuation. When only the bass sound effect is played back without playing the attention sound effect, while hearing the base sound effect, the impression of the hearing loss sound becomes stronger as the sound gets used to the sound, and there is a possibility that the hearing loss sound feels uncomfortable. For this reason, in addition to the base sound effect, an attention sound effect that attracts the listener's attention is reproduced, and control is performed to give a random impression to the sound effect. Thereby, it is possible to avoid an increase in the impression of the hearing-aid sound by getting used to the bass sound effect.

  In this way, by using two types of sound effects, the attention sound effect and the bass sound effect, it is possible to effectively reduce the uncomfortable feeling with respect to the hearing loss sound. Further, since the effect as a masking sound can be obtained by the attention sound effect and the bass sound effect, it is possible to obtain an effect of making it difficult to hear the content of the conversation as compared with the case where only the hearing-aid sound is used.

  Next, details of the attention sound effect data will be described. The attention sound effect is used for masking the conversation sound and for diverting the attention of the third party 3 from the conversation sound to the attention sound effect. When the third person 3 gets used to the sound, the effect of attracting attention is reduced, so that the reproduction of the attention effect sound is controlled so as to give a random impression so as not to get used to the sound.

  As shown in FIG. 6, attention sound effects having various time axis waveforms are used. The attention sound effect is a sound that decays after the sound pressure level rises in 2 seconds, for example, with different timbres such as bell, xylophone, and iron koto. There may be a case where one sound is played back during the two seconds, or a case where a plurality of two or more sounds are played back.

  To explain specifically, the attention sound effect of the koto tone, the koto sound may be heard only once in 2 seconds depending on the selected attention sound effect. In some cases, you can hear the sound of the height of the koto. For example, even when sounds are heard twice, the timing at which each sound is heard may be the same or different depending on the selected attention sound effect. That is, at least one of the tone color, the number of sounds, the pitch of each sound, and the timing at which each sound can be heard when each attention effect sound is reproduced, even if the attention effect sound is the same for 2 seconds. Yes.

  FIG. 7 is a diagram illustrating an example of the attention sound effect data 14 a stored in the storage unit 14. Thus, a plurality of sounds of various timbres are stored in the storage unit 14 as the attention sound effect data 14a. The sound effect control unit 13 randomly selects the attention sound effect data 14a to be reproduced from these.

  For example, the sound effect control unit 13 randomly selects three data from the attention sound effect data 14a having a reproduction time of 2 seconds. In this case, the three attention sound effect data 14a are reproduced over 6 seconds. However, as shown in FIG. 6, the sounds included in one attention sound effect data 14a are different from each other. For this reason, even when three attention sound effect data 14a for 2 seconds are selected, the number of sounds that can be heard in 6 seconds is a random number of 3 or more. Further, the tone color, the timing at which each included sound is reproduced, and the pitch of each sound differ depending on the selected attention sound effect data 14a. Further, each selected attention sound effect data 14a is reproduced by changing the volume level at random.

  In this way, a sound that receives a random impression is reproduced by randomly selecting a sound to be reproduced from among a plurality of attention effect sound data 14a consisting of various sounds and changing the volume level. be able to. As a result, the third party 3 who hears the attention sound does not get used to the sound, and the effect of attracting the listener's attention can be maintained.

  In the conversation protection system, not only a random impression but also a pleasant impression can be given to the person who hears the sound effect. When the attention sound effect data 14a is randomly selected and reproduced, depending on the attention sound effect data 14a to be continuously reproduced, a sense of incongruity may be felt. For example, if the pitch of the sound that is played continuously changes rapidly, the pitch is in a dissonant relationship, or the combination of timbres is not compatible, You may feel uncomfortable. For this reason, in the conversation protection system, a table in which the combination of the attention sound effect data 14a is stored in the sound effect output condition 14c of the storage unit 14. For example, the combination of the attention sound effect data 14a is set in the setting table in consideration of the combination of the timbre, the change in the pitch, the relationship between the pitches of the sounds that make up the consonance.

  FIG. 8 is an example of a setting table for the attention sound effect data 14a. In this setting table, combinations of timbres of attention sound effect data 14a, which are pleasant sounds when continuously reproduced, are set in advance.

  When using the setting table of the attention sound effect data 14a, the sound effect control unit 13 randomly selects a combination set in the setting table. Based on the selected combination, the attention sound effect data 14a of each tone color is selected at random. At this time, the sound effect control unit 13 also sets the volume level when reproducing each attention sound effect data 14a, but the sound volume level is also within a predetermined range set in advance so as not to feel uncomfortable. It is set up. Note that the volume level setting condition is also stored in the storage unit 14 as the sound effect output condition 14c, as in the setting table of the attention sound effect data 14a.

  For example, the combination of the selected attention sound effect data 14a is set to No. in the setting table shown in FIG. If it is 1, two attention sound effect data 14a for the bell tone and one attention sound effect data 14a for the xylophone tone are selected from the attention sound effect data 14a shown in FIG. Randomly selected. For example, when the attention sound effect data 14a selected at random is Bell B, Bell A, and Xylophone B, these three attention sound effect data 14a are sequentially displayed as shown in FIG. Played. In addition, for example, when the volume level is set to be changed between + 20% and −20% with respect to a predetermined level, and the randomly set volume levels are 90%, 100%, and 110% Each sound is played at this volume. That is, as shown in FIG. 9A, the sound of the bell B is reproduced at a volume level of 90% with respect to a predetermined volume level, the sound of the bell A is reproduced at a volume level of 100%, and 110%. The sound of xylophone B is reproduced at the volume level of.

  Note that the method of continuously playing a plurality of attention sound effects is not limited to a mode in which a plurality of sounds are continuously played as shown in FIG. 9A, and each sound is played as shown in FIG. You may reproduce | regenerate so that a part of may overlap. In this case, the overlapping of each sound, that is, the reproduction timing of each sound may be set at random within a predetermined range set in advance.

  Further, the attention sound effect selection method may be selected reflecting the user's preference. For example, the combination of timbres in the setting table shown in FIG. 8 may be set according to the user's preference, or the attention sound effect data 14a is selected from the data shown in FIG. 7 according to the user's preference. The setting table may be a combination of these at random. Further, the attention sound effect data 14a that can be selected when the sound effect control unit 13 selects data randomly is stored in advance in the user's preference without changing the setting table of FIG. 8 set irrespective of the user's preference. It is also possible to reflect the user's preference by narrowing down according to the above.

  Next, the reproduction process of the bass sound effect data 14b and the attention sound effect data 14a will be described. FIG. 10 is a flowchart showing a process when reproducing the bass sound effect data 14b. FIG. 11 is a flowchart showing a process when reproducing the attention sound effect data 14a.

  First, a process for reproducing the bass sound effect data 14b will be described. The sound pressure waveform generated by the input sound analysis unit 11 from the conversational sound collected by the microphone 20 is monitored by the output sound control unit 15 (step S1 and step S1; No).

  When the sound pressure level obtained from the conversational sound exceeds the preset threshold B for the base sound effect (step S1; Yes), the sound pressure level obtained from the conversation sound is based on the setting in the sound effect output condition 14c. The selected bass sound effect data 14b is reproduced (step S2). The output sound control unit 15 continues to monitor the sound pressure level (Step S3 and Step S3; No).

  When the sound pressure level obtained from the conversational voice is below the threshold value B (step S3; Yes), the reproduction of the bass sound effect data 14b is stopped (step S4). At this time, the bass sound effect is stopped to fade out so that the third party does not feel uncomfortable. Even after the reproduction of the base sound effect data 14b is stopped, the output sound control unit 15 continues to monitor the sound pressure level obtained from the conversational sound, and controls the reproduction and stop of the base sound effect data 14b. That is, as shown in FIG. 5, while the sound pressure level of the conversation voice exceeds the predetermined threshold value B, the base sound effect data 14b is reproduced.

  If the attention sound effect data 14a is reproduced when the reproduction of the bass sound effect data 14b is stopped, the reproduction of the attention sound effect data 14a is also stopped in the same manner as the base sound effect data 14b. You may control to. Further, the bass sound effect data 14b may be in a mode in which the playback sound volume is controlled while always playing back the base sound effect data 14b, in addition to a mode in which playback and stop are controlled. Specifically, when the sound pressure level of the conversation voice exceeds the threshold value B, the playback volume of the base sound effect data 14b is faded in, and the playback is continued while maintaining the volume when the predetermined volume is reached. When the sound pressure level of the conversational voice falls below the threshold value B, the volume may be reduced so as to fade out.

  Next, a process for reproducing the attention sound effect data 14a will be described. The output sound control unit 15 monitors the sound pressure waveform obtained from the conversational voice as in the case of the bass sound effect data 14b (Step S11 and Step S11; No).

  When the sound pressure level obtained from the conversational voice exceeds a preset threshold value A for the attention sound effect (step S11; Yes), the sound effect control unit 13 causes the attention sound effect data to be obtained. 14a is selected at random (step S12). Furthermore, the sound effect control unit 13 randomly sets the volume level when reproducing the attention sound effect data 14a within a predetermined range (step S13).

  And the output sound control part 15 reproduces | regenerates the attention sound effect data 14a selected at random from the speaker 30 with the sound volume set at random (step S14). The output sound control unit 15 continues to monitor the sound pressure level during these processes, and when the sound pressure level obtained from the conversational sound exceeds the threshold A, the next attention effect is obtained. The sound data 14a is reproduced. That is, as shown in FIG. 5, each time the sound pressure level of the conversational voice exceeds a predetermined threshold A, the randomly selected attention sound effect data 14a is reproduced at a randomly set volume level. The

  In the present embodiment, an aspect using the setting table of the attention sound effect data 14a shown in FIG. 8 is shown, but the present invention is not limited to a single setting table, and an aspect using a plurality of setting tables. There may be.

  For example, a plurality of setting tables may be prepared so as to include a lot of attention sound effects of a specific tone color, and the setting tables may be selected and used according to the conversation voice. Specifically, a table containing a lot of attention sound effects such as xylophone, which receives a quiet impression when the sound pressure level is low, matches the sound pressure level of the conversational voice, and when the sound pressure level is high Use a table that contains a lot of attention-sounding timbres that receive a strong impression.

  Moreover, when the input sound analysis unit 11 analyzes the conversational voice, the voice quality and the sex may be determined, and the setting table may be selected and used based on the determination result. For example, the setting table is set and used in advance so that the attention sound effect data 14a having a high effect of masking the conversational sound and reducing the sense of incongruity of the hearing loss sound is selected according to voice quality and gender. Specifically, in accordance with the voice quality of conversational voice, for example, using a setting table that contains many attention sound effects such as iron koto for female high voice, xylophone for male low voice, etc. Use a setting table that contains a lot of attention sound effects.

  As described above, if the setting table is used so that the tone color and the pitch of the attention sound data 14a are selected in accordance with characteristics such as the voice quality of the conversation voice while the attention sound effect data 14a is selected at random, the conversation voice can be used. It is possible to further increase the masking effect on the sound and the effect of reducing the sense of incongruity of the hearing loss.

  Note that the volume level when the attention sound effect data 14a is reproduced may be changed in accordance with the conversational voice. Further, the volume level at the time of selection and reproduction of the base sound effect data 14b may be set according to the conversational sound.

  In this embodiment, the combination of attention sound effects is set in the setting table in consideration of the timbre, but in addition to this, the attention sound effects according to the sound and music used as the base sound effects are shown. It is also possible to adopt a mode in which these combinations are preset. More specifically, a combination of attention sound effects using the chirping of a bird is created as a setting table, and this setting table is used in association with a bass sound effect consisting of the sound of a river. As a result, when a river murmur sound is selected as the base sound effect data 14b, it is possible to reproduce the bird's song as the attention sound effect data 14a. In this way, a comfortable impression can be given to the listener by setting so as to use an attention sound effect that matches the base sound effect.

  Further, in the present embodiment, as shown in FIG. 6, a mode is shown in which three data are selected and reproduced from the attention sound effect data 14a having the same reproduction length. However, the present embodiment is not limited to this, and the number of attention effect sound data 14a to be selected and the length of each attention effect sound data 14a may be different.

  Specifically, for example, an attention sound effect having a length ta shown in FIG. 12A is generated by combining two attention effect sound data 14a having the same length t1 as shown in FIG. 12B. Alternatively, as shown in FIG. 5C, it may be generated by combining three attention sound effect data 14a having different lengths. For attention sound effects, if you can draw attention while giving a random impression so that the effect of attracting attention by third party 3 is not lost, tone, pitch, volume, playback timing The generation method and the like are not limited to the examples described in this embodiment.

  As described above, according to the present embodiment, two kinds of sounds, the attention sound effect data 14a and the base sound effect data 14b, are reproduced according to the volume of the conversational sound to be protected. Can make it difficult to hear the content of the conversation.

  Further, by reproducing the attention sound effect data 14a and the base sound effect data 14b in addition to the hearing-aid sound that makes it difficult to hear the conversational sound, the conversation is effectively protected and the sense of incongruity with the hearing-aid sound is reduced. Can do.

  In addition, the attention sound effect data 14a is reproduced while changing the tone color, pitch, volume, etc., thereby giving a random impression to the listener, and getting used to the reproduced sound to reduce the sense of incongruity with the hearing loss sound. Can be prevented from fading.

  As described above, the present invention protects the content of a conversation by playing the sound so that the conversation voice is not heard by a third party, and prevents the sound played by the third party from feeling uncomfortable or uncomfortable. This is a useful technique.

DESCRIPTION OF SYMBOLS 10 Sound processing apparatus 11 Input sound analysis part 12 Hearing loss sound generation part 13 Sound effect control part 14 Storage part 14a Attention sound effect data 14b Base effect sound data 14c Sound effect output condition 15 Output sound control part 20 Microphone 30 Speaker 40 Output sound operation Department 50 Table 51 Partition 52 Waiting seat

Claims (12)

A microphone to collect the conversational voice,
An attention effect sound that shows a time axis waveform that decays after the sound pressure level rises within a few seconds, and a base effect sound that shows a time axis waveform whose sound pressure level changes more slowly than the attention effect sound are stored. A storage unit;
A speaker for playing one or both of the attention sound effect and the base sound effect toward a third party excluding the conversation party performing the conversation;
Control that reproduces the base sound effect by the speaker at least while the sound pressure level of the conversation voice collected by the microphone exceeds a first threshold value, and the sound pressure level of the conversation voice is a second And a control unit that controls one or both of the control to reproduce the attention sound effect every time the threshold value is exceeded.   The control unit stops reproduction of one or both of the base sound effect and the attention sound effect by the speaker when the sound pressure level of the conversational sound falls below a predetermined threshold value. The conversation protection system according to claim 1.   Each time the control unit reproduces the attention sound effect, at least one of the number of sounds that can be heard within a predetermined time, the timing at which the sound is heard, the tone color of the sound, and the pitch of the sound changes. The conversation protection system according to claim 1, wherein reproduction of the attention sound effect is controlled.   The conversation control system according to claim 1, wherein the control unit generates the attention sound effect using a sine wave. A masking sound generator for generating a masking sound that masks the sound and makes it difficult to hear based on frequency characteristics of the sound collected by the microphone;
The conversation control system according to claim 1, wherein the control unit reproduces the masking sound generated by the masking sound generation unit.   The masking sound generation unit extracts a spectral envelope and a spectral fine structure from the sound collected by the microphone, and sets an inversion axis for switching the spectral envelope up and down around an axis extending in the frequency direction, Generating a modified spectral envelope by reversing the spectral envelope by reversing the spectral envelope around the inversion axis, and generating hearing loss sound that combines the modified spectral envelope and the spectral fine structure; The conversation protection system according to claim 5, wherein the masking sound is used. The storage unit stores a plurality of the attention sound effects,
The conversation control system according to claim 1, wherein the control unit reproduces the attention sound effect selected at random from the storage unit. In the storage unit, a combination of a plurality of attention sound effects set based on the tone color and pitch of each attention sound effect is stored as a setting table,
8. The conversation protection system according to claim 7, wherein the control unit reproduces an attention sound effect based on a combination randomly selected from the setting table.   The conversation control system according to claim 1, wherein the control unit randomly changes a volume at the time of reproducing each attention sound effect.   The conversation protection system according to claim 1, wherein the attention sound effect is a sound of a musical instrument. A voice collection step that collects the conversational voice;
A base that reproduces a base sound effect that exhibits a time-axis waveform in which the sound pressure level gradually changes while the sound pressure level of the conversational sound collected in the sound collecting step exceeds the first threshold value. Time to decay after the sound pressure level rises for several seconds when the sound pressure level of the conversational sound collected in the sound effect reproducing step and the sound collecting step exceeds the second threshold value And a sound effect reproduction step including any one or both of the attention sound effect reproduction steps for reproducing the attention sound effect indicating the axis waveform. The effect of stopping the reproduction of one or both of the base sound effect and the attention sound effect when the sound pressure level of the conversational sound collected in the sound collecting step falls below a predetermined threshold value. The speech protection method according to claim 11, further comprising a sound stop step.

Images