JP2839983B2 - Pseudo sound generator - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a prosthetic device for generating a sound equivalent to that of a healthy person in a human body from which a vocal cord or other vocal organs have been removed or congenitally lacked.

[0002]

2. Description of the Related Art Generally, a human utters a voice by generating a strong airflow from an airway in a narrow gap of a vocal cord. The volume and tone are determined by the pressure of the air, the interval between the vocal cords, and the like.
Furthermore, the timbre is changed by changing the shape of the oral cavity and the nasal cavity at the same time, and the language is pronounced in combination with the sound generated in the oral cavity. When some parts of the vocal organs, especially the vocal cords and surrounding tissues, are removed from the human body for any reason,
Thereafter, communication by voice equivalent to a healthy person cannot be performed, and a malfunction of a voice recognition device or the like is likely to occur. Reasons for such acquired vocal organ removal include malignant laryngeal cancer, extreme laryngitis and trauma. Such disorders also occur in the human body, which naturally lacks such organs. For those who lack a part of the vocal organs in this way, they often have to take other means of communication other than voice, such as using sign language, writing, or using a keyboard. Press the vibrator from outside the body,
At the same time, the method of generating sound by deforming the larynx, oral cavity, and nasal cavity, and vibrating tissues other than the original vocal organs, such as the esophagus, with air flowing backward from the breathing or stomach, and simultaneously the larynx, oral cavity, and nasal cavity The original operation of the vocal organs had to be replaced by a method of deforming and vocalizing (esophageal vocalization), a method of attaching a laryngeal flute to the back of the head, and generating a voice.

[0003]

When the vibrator is used in some way to replace the operation of the original vocal organ, a device such as a vibrator must be carried at all times, and special training is required. There is a problem that. The problem with the esophageal vocalization method is that it requires more special training, may cause great mental and physical distress, and may not be possible depending on the situation of lack of vocal organs. There is. In addition, in any of these methods, the method using the laryngeal whistle also has a problem that the clarity of the uttered voice is significantly worse than that of a healthy person.

[0004] It is an object of the present invention to be used as a prosthesis,
It is an object of the present invention to provide a pseudo-sound generating device capable of generating a clear sound without giving a pain to a wearer who lacks a part of a vocal organ.

[0005]

According to a first aspect of the present invention, there is provided a pseudo-speech generating apparatus comprising: detecting means for detecting a state of a vocal organ of a human body; a detection amount detected by the detecting means; Voice generating means for simulating the sound to be generated based on the detected amount.

[0006]

Since there are detecting means for detecting the state of the vocal organs and vocal means for simulating the sound to be generated by the vocal organs based on the amount detected by the detecting means, Even if a part of the organ is missing and cannot actually be uttered, the voice to be uttered can be generated in a pseudo manner, and communication by voice can be performed.

In this case, the detecting means includes a pharynx, a larynx,
With respect to at least one or more organs of the airway, face, oral cavity, and nasal cavity, at least one or more state quantities of muscle state, nerve state, displacement amount, and stress of the organ are detected. Good. This is because the voice to be uttered is determined by the state of the muscles and nerves of the pharynx, larynx, airway, face, oral cavity, and nasal cavity, and the displacement and stress of these organs. By causing the utterance means to simulate the sound based on the detection amounts representing the states of these organs, it is possible to generate the sound intended by the person wearing the pseudo sound generator. . In general, the types of sounds (pronunciation, tone, volume, voice characteristics, etc.) generated by a normal speaker are limited, so it is not necessary to measure all of the state quantities of all of these organs. The organ to be measured and the state quantity of the measurement target may be determined based on the requirement for the quality of the pseudo sound to be generated.

[0008] The utterance means searches the database section based on a detected waveform to be generated corresponding to the detected amount and a database section based on the input detected amount, and generates a voice based on the search result. It is preferable to be configured with voice conversion means for causing the sound to be converted. Based on the detected amount of the state of each organ such as the pharynx, larynx, airway, face, oral cavity, nasal cavity, etc., the sound waveform to be generated assuming a model of each organ is obtained by hydrodynamic calculation It is conceivable, however, that such calculations require enormous computational power and are not practical. Since the types of voices are limited as described above, a database unit storing the correlation between these types of voices and the corresponding detection amounts is prepared in advance, and the detection amounts representing the state of each organ are input. If the database unit is searched each time, the pseudo voice generating apparatus can be configured with a small amount of hardware and software. At this time, if the database unit can be learned and updated by an external command, it is possible to generate a sound most suitable for each wearer.
This also means that for those who acquire some of the vocal organs acquired,
By recording the actual voice before losing and constructing the contents of the database unit based on this, it is possible to eliminate a sense of discomfort to the voice before and after losing the vocal organ.

In the present invention, the detection means has a size of typically about 5 mm square or less, and is used by being embedded in each organ. On the other hand, the vocalization means is provided outside the wearer's body, particularly on the neck, collar, chest and the like. Both the detection means and the utterance means are formed extremely finely.

[0010]

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

[First Embodiment] FIG. 1 is a diagram showing an arrangement of sensor chips in a pseudo sound generator according to one embodiment of the present invention.
FIG. 2 is a block diagram showing a configuration of a sensor chip, and FIG. 3 is a block diagram showing a configuration of a voice synthesizing unit in this embodiment.

This pseudo-speech generating apparatus comprises five sensor chips 101 to 105 as detecting means and a voice synthesizing unit as voicing means. As shown in FIG. 1, the first sensor chip 101 is provided on the pharynx to detect the state of the pharyngeal muscle, and the second sensor chip 102 is provided on the larynx to detect the state of the airway 131. The remaining three sensor chips 103 to 105 are provided in the oral cavity 132 to detect the state of the oral cavity 132, and are mainly for the cheek state, the tongue state, and the opening angle of the oral cavity, respectively. Are respectively detected. These sensor chips 10
1 to 105 are fixed to the organ to be detected by being adhered or embedded in the skin, mucous membrane or cellular tissue.

Next, the configuration of each of the sensor chips 101 to 105 will be described. Each sensor chip 101 to 105
Are the sensor units 111 and 11 as shown in FIG.
A signal converter 112 for amplifying the weak signal detected in step 1 and converting it into a modulated signal; a chip monitoring unit 113 for monitoring the state in the sensor chip and outputting it as a monitor signal; And a signal transmission unit 114 for transmitting the signal to The sensor unit 111 detects a mechanical displacement, and includes, for example, a mechanical amount detecting element using a piezoelectric effect, a strain gauge, and the like. In addition, the sensor unit 111 may be a unit that directly extracts a signal from a muscle, a nerve, or the like using a stylus or the like. As will be described later, the distance between the speech synthesis unit and each sensor chip is at most 10 cm or less.
The transmission power of the radio wave in 4 may be very weak.
The carrier frequency of the radio wave from the signal transmitting unit 114 is different for each of the sensor chips 101 to 105.

Although the operation of the sensor chips 101 to 105 requires power, the transmission power is extremely weak as described above, and the power consumption of the entire sensor chip can be reduced by using currently available power saving techniques. Since the power can be suppressed to an extremely small level, this power can be supplied from a minute primary battery provided in the sensor chip. In this case, the primary battery can be used virtually semi-permanently. Alternatively, the sensor chip may be radiated from the outside with radio waves at a level that does not affect the living body, and the radio waves may be rectified and stored in a secondary battery provided in the sensor chip. At this time, the amount of radiation of the power supply radio wave should be minimized, so it is necessary to know the charge amount of the secondary battery from the outside, and the above-described monitoring unit 113 monitors the charge amount of the secondary battery. What should I do?
In addition, power may be supplied using an element that converts acceleration applied to the sensor chip with vibration or movement of a human body into electric energy.

On the other hand, the voice synthesizing unit includes a receiving unit 12 for receiving signal radio waves from each of the sensor chips 101 to 105.
1. Rewritable memory 1 constituting a database unit
22, a microcomputer 123 and a microcomputer 12 which search the memory 122 based on signal radio waves received by the receiving unit 121 and determine the type of sound to be generated
Speech synthesizer 124 that generates a speech waveform based on the speech type determined in step 3, and amplifier 12 that amplifies the speech waveform
5. Speaker 126 connected to the output side of amplifier 125
Consists of The voice synthesizing unit is mounted outside the wearer's body, for example, on the neck or collar. When a voice synthesizing unit is provided on the neck or collar, each of the sensor chips 101 to 105
The distance between the voice signal and the voice synthesis unit can be made approximately 10 cm or less, and the transmission power of the sensor chips 101 to 105 can be extremely small. Each sensor chip 1
When 01 to 105 are supplied by radio waves, a transmission unit for generating power supply radio waves may be provided in the voice synthesizing unit.

In the memory 122 as a database unit,
A database that describes the correlation between the state of each vocal organ and the voice is stored. Also, in the memory 122,
An operation panel 127 provided outside can be connected. The operation panel 127 is for learning and updating the database stored in the memory 122 at any time, and can input a volume, a tone, general phonetic symbols, and the like. The wearer operates the operation panel 127 to change what kind of sound while putting emphasis on each vocal organ in the same manner as actually uttering, that is, changing the state of each vocal organ according to the voice to be generated. By instructing whether to generate, the database is learned and updated, and the type of voice generated by this voice synthesizing unit when the emphasis is made as described above, that is, pronunciation, pitch, tone, timbre , Volume and voice characteristics can be changed freely.

Next, the operation of the pseudo sound generator will be described.

When the wearer applies a force to each of the vocal organs in response to the sound to be generated, displacement occurs in these vocal organs, and the sensor units 11 of the sensor chips 101 to 105 are displaced.
At 1 the displacement is detected. When the sensor unit 111 directly extracts a signal from a nerve or a muscle, a signal corresponding to how to apply force to each vocal organ is directly detected. Then, the signal detected by the sensor unit 111 is converted into a modulation signal by the signal conversion unit 112 and transmitted as a radio wave from the signal transmission unit 114. The sensor chip 101 arranged on the pharynx mainly detects information on displacement of the vocal cord muscles, and the sensor chip 102 arranged on the larynx mainly detects information on displacement due to the pressure of air from the lungs. Sensor chips 103 to 105 arranged
Detects mainly the displacement of each part of the oral cavity.

Signal radio waves from each of the sensor chips 101 to 105 are received and demodulated by the receiving section 121 of the voice synthesizing unit. Since the carrier frequency of the signal wave differs for each sensor chip, it is easy to identify which sensor chip the signal wave is coming from. The demodulated signal is the detected amount of each of the sensor chips 101 to 105. The microcomputer 123 searches the database in the memory 122 for the type of sound corresponding to the detected amount, and finds the closest sound pattern. Is extracted, and the type of sound to be generated is determined. The voice synthesizer 124 generates a voice waveform based on the determined voice type, and the voice waveform is amplified by the amplifier 125 and sent out from the speaker 126 into the air as an actual sound wave. As described above, the database can be learned and updated, and the detected amount when a certain force is applied to each vocal organ and the sound to be generated when the specific force is applied are determined. Since it is stored in this database, the type of sound that the wearer intends to generate can be searched without a doubt, and therefore, the sound that the wearer intends to generate can be generated accurately. In addition, since voice is generated based on the type of voice stored in the database in advance, clear voice can be obtained.

[Embodiment 2] FIG. 4 is a diagram showing an arrangement of marker chips in a pseudo sound generator according to another embodiment of the present invention, FIG. 5 is a block diagram showing a configuration of the marker chips, and FIG.
FIG. 3 is a block diagram showing a configuration of a speech synthesis unit in this embodiment. In this embodiment, a voice is generated by detecting a displacement between the respective vocal organs.

This pseudo speech generator comprises five marker chips 201 to 205 and a speech synthesis unit. As shown in FIG. 4, each of the marker chips 201 to 205 includes a pharynx, a laryngeal airway 131 side, a cheek,
It is provided on the tongue and chin. These marker chips 201 to
205 is fixed to the organ to be detected by adhering or embedding it to skin, mucous membrane or cellular tissue. As shown in FIG. 5, each of the marker chips 201 to 205 has an oscillation unit 2 that generates a high frequency of, for example, about several hundred MHz.
11 and a transmission unit 212 that outputs the high frequency generated by the oscillation unit 211 as a radio wave. The oscillating frequency of the oscillating unit 211 is different for each of the marker chips 201 to 205. Marker chips 201-205
Unlike the sensor chips 101 to 105 of the first embodiment described above, the sensor chip does not have a sensor unit for detecting displacement or the like, and continuously transmits a high frequency having a constant frequency. Therefore, power consumption can be further reduced as compared with the above-described sensor chip. These marker chips 201 to 20
5 is supplied to the sensor chips 101 to 105 described above.
Is the same as

On the other hand, as shown in FIG. 6, the voice synthesizing unit includes three receiving units 221 to 223 for receiving radio waves from the respective marker chips 201 to 205, and each of the receiving units 221 to 221.
23, a phase comparator 224 for comparing the phases of the radio waves received by each other, a rewritable memory 122 constituting a database unit, and the marker chips 201 to 205 based on the phase comparison values obtained by the phase comparator 224. A microcomputer 225 for calculating relative position fluctuations and searching the memory 122 to determine the type of voice to be generated; a voice synthesis unit 124 for generating a voice waveform based on the type of voice determined by the microcomputer 225; An amplifier 125 amplifies the waveform, and a speaker 126 connected to the output side of the amplifier 125. Memory 122, speech synthesizer 1
24, the amplifier 125, and the speaker 126 are the same as those in the first embodiment.
The same as in. The operation panel 127 can be connected to the memory 122, and the database stored in the memory 122 can be learned and updated, as in the first embodiment. This speech synthesis unit is worn outside the body of the wearer, for example, on the neck or collar. At this time, the three receiving units 221 to 223 are arranged so as not to be aligned on the same straight line.

Next, the operation of the pseudo sound generator will be described.

When a force is applied to each of the vocal organs in accordance with the sound to be generated by the wearer, displacement occurs in these vocal organs, and the positions of the marker chips 201 to 205 change.
Each marker chip 201-205 and each receiving section 221-222
3 changes, and the phases of the radio waves received by the receivers 201 to 203 are shifted. Since the three receiving units 221 to 223 of the voice synthesizing unit are not on the same straight line, the phase comparator 214 detects a change in the phase difference of the radio wave received by each of the receiving units 221 to 223, so that each marker The moving direction and the moving distance of the chips 201 to 205 are obtained, and these marker chips 201 to 205 are obtained.
The displacement amount of the organ provided with is determined. In order to distinguish between movement of the entire head due to movement of the neck and the like and movement of each organ, position variations of all the marker chips 201 to 205 are obtained, and relative position movement between the marker chips, particularly Oral cavity 132
(Cheek, tongue, jaw) movement is detected. Since the frequency of the radio wave transmitted from each of the marker chips 201 to 205 is different, it is easy to identify the radio wave transmitted from each of the marker chips.
The displacement amounts of 01 to 205 can be detected. With respect to the displacement amounts of the respective organs obtained in this manner, similarly to the first embodiment, a database in which a correlation between the displacement amounts of the respective organs and the sound to be generated is described is constructed in the memory 122 in advance. By doing so, the sound that the wearer intends to generate can be generated accurately.

Although the embodiments of the present invention have been described above, for example, a sensor chip is used to detect the state of the airway in consideration of the accuracy of the sound to be generated, the size of the portion to be embedded in the body, and other factors. It is also effective to use a combination of the above-described sensor chip and marker chip, such as using a marker chip in a part.

The features of the pseudo sound generator according to the present invention are as follows.
A clear voice can be generated, and a burden such as special training is not required for the wearer. In the present invention, in particular, the sensor chip, the marker chip, and the like must be extremely finely configured. This is realized by a micromachining technique using semiconductor manufacturing technology.

[0027]

As described above, the present invention provides a detecting means for detecting the state of a vocal organ, and a vocal sound generated quasi-generated based on the amount of sound to be generated by the vocal organ based on the detection amount of the detecting means. By providing the means, even in the case where a part of the vocal organs cannot be actually uttered due to lack of vocal organs, it is possible to perform clear voice communication without performing special training, and This has the effect of reducing malfunctions of the recognition device and the like and reducing the mental burden on the wearer.

[Brief description of the drawings]

FIG. 1 is a diagram showing an arrangement of sensor chips in a pseudo sound generation device according to a first embodiment.

FIG. 2 is a block diagram illustrating a configuration of a sensor chip according to the first embodiment.

FIG. 3 is a block diagram illustrating a configuration of a speech synthesis unit according to the first embodiment.

FIG. 4 is a diagram illustrating an arrangement of marker chips in a pseudo sound generator according to a second embodiment.

FIG. 5 is a block diagram illustrating a configuration of a marker chip according to a second embodiment.

FIG. 6 is a block diagram illustrating a configuration of a speech synthesis unit according to a second embodiment.

[Explanation of symbols]

 101 to 105 Sensor chip 111 Sensor unit 112 Signal conversion unit 113 In-chip monitoring unit 114 Signal transmission unit 121, 221-223 Receiving unit 122 Memory 123, 225 Microcomputer 124 Voice synthesis unit 125 Amplifier 126 Speaker 127 Operation panel 201-205 Marker Chip 211 Oscillator 212 Transmitter 224 Phase comparator

Claims (5)

(57) [Claims] 1. A detecting means for detecting a state of a vocal organ of a human body, and a detection amount detected by the detecting means is inputted, and a sound to be generated by the vocal organ is simulated based on the detected amount. A pseudo-speech generating device having a vocalization means for generating a pseudo sound. 2. The method according to claim 1, wherein the detecting means comprises at least one of a pharynx, a larynx, an airway, a face, an oral cavity, and a nasal cavity, at least one of a muscle state, a nerve state, a displacement amount, and a stress of the organ. 2. The pseudo-speech generating apparatus according to claim 1, wherein the pseudo-sound generating apparatus detects at least three kinds of state quantities. 3. The human body according to claim 1, wherein the detecting means is embedded inside the human body, and the vocal means is provided outside the human body.
A pseudo-speech generating device according to claim 1. 4. A vocalization means searches a database section in which a waveform of a voice to be generated corresponding to the detected amount is recorded, and searches the database section based on the input detected amount, and generates a voice based on the search result. 4. The pseudo-sound generating apparatus according to claim 1, wherein said pseudo-sound generating apparatus comprises a sound converting means for causing the pseudo sound to be generated. 5. The pseudo-speech generating apparatus according to claim 4, wherein the database unit is capable of learning and updating by a command from the outside.