WO2011022314A1

WO2011022314A1 - Method and apparatus for increasing voice loudness

Info

Publication number: WO2011022314A1
Application number: PCT/US2010/045568
Authority: WO
Inventors: Jessica E. Huber; Scott Kepner; Derek Tully; James Thomas Jones; Kirk Solon Foster
Original assignee: Purdue Research Foundation
Priority date: 2009-08-17
Filing date: 2010-08-16
Publication date: 2011-02-24

Abstract

A method and system (10) for improving communication. The system (10) comprises an assembly fixedly positioned during use in close proximity to a user's ear. The assembly includes an accelerometer (14) to detect the initiation and duration of the user's speech and an output presentation system. The output presentation system comprises a single non- occlusive ear fitting (16) that presents noise (31, 68) that is unrelated to the sound- frequency or intonation, of the user's speech. The system (10) further comprises a control arrangement (38) to maintain presentation of the noise (31, 68) throughout the detected duration of the user's speech for a prescribed period of time.

Description

TITLE: METHOD AND APPARATUS FOR INCREASING VOICE LOUDNESS

CROSS REFERENCES TO RELATED APPLICATIONS

[0001] The following application claims priority to co-pending U.S. Provisional

Patent Application Serial No. 61/234,401 filed August 17, 2009 entitled METHOD AND APPARATUS FOR INCREASING VOICE LOUDNESS. The above-identified application is incorporated herein by reference in its entirety for all purposes.

GOVERNMENT FUNDING

[0002] This invention was made with government support under National Institutes of

Health ("NIH") Grant No. R01DC009409. The United States government has certain rights in the invention.

TECHNICAL FIELD

[0003] The present disclosure relates to a method and system for increasing voice loudness, and more specifically, the present disclosure comprises a voice enhancement system and method used to elicit the Lombard effect. The Lombard effect is an external cue for increasing voice loudness. The present method and system uses the Lombard effect to assist individuals experiencing problems with vocal intensity, articulation, and/or volume as a result of a physically degenerative condition.

BACKGROUND

[0004] Parkinson's Disease is a progressive movement disorder in which there is a deficit in dopamine production in the basal ganglia. Parkinson's Disease is a common degenerative disease of neurological origin. Parkinson's Disease can be characterized by reduced amplitude and velocity of movement and difficulty with movement initiation and premotor programming. Deficits in speech production have been reported in individuals with Parkinson's Disease that include both voice and articulation problems. Such deficiencies in the speech of individuals with Parkinson's Disease are characterized by hypophonia (reduced loudness), monoloudness, monopitch, disordered rate and articulation, and a voice that is hoarse, breathy, harsh, and/or tremulous.

[0005] Respiratory-related deficits in individuals with Parkinson's Disease have been demonstrated by reductions in forced vital capacity and forced expiratory volume in one second, even under the effects of anti-Parkinsonian medications. Respiratory kinematic patterns for speech are also altered in individuals with Parkinson's Disease. Individuals with Parkinson's Disease may experience respiratory and/or laryngeal muscle weakness or changes to the neural drive to these muscles. Reduced vocal fold closure, reduced respiratory support, respiratory and laryngeal muscle weakness, and changes in neural drive to these muscles are all potential causes of hypophonia, a common symptom associated with Parkinson's Disease.

[0006] There is evidence that hypophonia in individuals with Parkinson's Disease is not only due to disease-related physiologic limitations, such as muscle rigidity or weakness. Additionally, coordination of movements across subsystems and between articulators, which is critical for the production of speech, may be deficient in individuals with Parkinson's Disease.

[0007] There is belief in the medical field that one of the contributing factors to

Parkinson's Disease-associated movement problems, including hypophonia, is sensory deficits. Due to sensory deficits, individuals with Parkinson's Disease may have an impaired perception of their own speech volume, resulting in inaccurate internal representations of movement targets and vocal intensity targets.

[0008] Frontal lobe and/or executive impairments may cause patients with Parkinson's

Disease to have problems developing their own plan of action or initiating and maintaining goal- directed behavior, along with concept formation and self-monitoring behavior. In all, persons with Parkinson's Disease typically have problems with efficient strategy formation. Not only are these underlying cognitive and behavioral deficits likely to contribute to speech and language difficulties, but also to the lack of success of treatment strategies for persons with Parkinson's Disease.

[0009] Parkinson's Disease has been identified as just one illness that can result in impairments relating to speech production. Other illnesses or diseases are also known to exist that have similar symptoms of low speech loudness and/or respiratory deficiencies. There is some disbelief in the medical field that long-term improvement of communication in Parkinson's Disease patients (PDPs) is possible through the use of a device that initiates an external cue for increasing voice loudness. For example, in a reviewer's comment to an NIH proposal relating to the possibility of conditioning the patients to achieve a long-term cure to speech related impairments using a device that initiates an external cue for increasing voice loudness, it was generally stated that "[i]t is impossible to determine if patients will stick with the protocol" and that the proposed conditioning of patients could be "too irritating or distracting, in which case...[the proposal] could actually turn out to be a disincentive for talking."

[0010] Further evidence supporting the medical field's disposition that therapeutic and conditioning of PDP' s communication or speech levels using a device that initiates an external cue for increasing voice loudness is not possible was found throughout the reviewer's comments in applicant's NIH proposal relating to the subject application. Such comments by the NIH reviewer included, for example, the following general statements: "there is no real demonstration that [the proposal] is likely to work under the conditions described"; and "[t]o date, the Lombard effect has never been harnessed in this manner, and [the proposal used] to achieve LSVT-type goals is also original" (LSVT is a traditional behavioral speech therapy for PDPs). The NIH reviewer's comments to applicant's proposal went so far as to surmise that such approaches suggested therein teach away from the prior art. More specifically, the NIH reviewer's comments stated that when employing the proposed approaches for conditioning PDPs, "[c]autions are taken to consider potential issues with adaptation as well as hearing safety".

SUMMARY

[0011] One example embodiment of the present disclosure comprises a system to improve communication having an arrangement fixedly positioned during use in close proximity to a user's ear. The arrangement comprises an accelerometer to detect the initiation and duration of the user's speech and an output presentation system. The output presentation system comprises a non-occlusive ear fitting that presents noise that is unrelated to the sound-frequency or intonation of the user's speech. The system further comprises a control mechanism to maintain presentation of the noise throughout the detected duration of the user's speech for a prescribed period of time.

[0012] Another example embodiment of the present disclosure comprises a method to improve communication in a patient with Parkinson's Disease comprising training the patient's speech production mechanism and/or speech motor control mechanisms through a voice enhancement system that introduces noise to patient during their speech.

[0013] Yet another example embodiment of the present disclosure comprises a system for training a patient with Parkinson's Disease to have improved communication. The system comprises an activation device responsive to movement or vibrations in a patient during speech and upon sensing the patient's speech movement or vibration produces an output signal. The system also comprises a presentation assembly having a non-occlusive ear fitting that transmits noise to the ear of a patient during a noise enabling conditions. The system also comprises a control arrangement coupled to the activation device for receiving the output signal. The control arrangement comprises a signal processor having a prescribed threshold that is analyzed against the output signal such that one or more output signals above the threshold generates the noise enabling condition resulting in noise transmission from the control arrangement to the presentation assembly.

[0014] A further example embodiment of the present disclosure comprises a method for training a patient to have increased voice loudness. The method comprises positioning a non- occlusive ear fitting in a patient's ear that transmits noise to the ear during a noise enabling condition. The method further comprises positioning externally an accelerometer on a patient at a location that moves or vibrates during speech. The method also comprises transmitting an output signal from the accelerometer to a control arrangement upon sensing movement or vibration experienced during speech of the patient. The method yet also comprises analyzing the output signal with a signal processor within the control arrangement against a prescribed threshold and transmitting audio noise to the ear fitting from the control arrangement when the output signal is above the threshold generating the noise enabling condition.

BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The foregoing and other features and advantages of the present disclosure will become apparent to one skilled in the art to which the present disclosure relates upon consideration of the following description of the invention with reference to the accompanying drawings, wherein like reference numerals, unless otherwise described refer to like parts throughout the drawings and in which: [0016] FIG. 1 is a voice enhancement system constructed in accordance with one embodiment of the present disclosure;

[0017] FIG. 2A is an earpiece positioned within an ear of a patient using the voice enhancement system and methods for treatment and the positioning of an activation device in accordance with one example embodiment of the present disclosure;

[0018] FIG. 2B illustrates another example embodiment for the positioning of an activation device for a voice enhancement system of the present disclosure;

[0019] FIG. 2C illustrates an example embodiment of a voice enhancement system comprising a compact design where the entire system is positioned near the patient's ear;

[0020] FIG. 3 is a rear cover of a housing for the voice enhancement system illustrated in FIG. 1;

[0021] FIG. 4 is a block diagram illustrating a control arrangement of a voice enhancement system constructed in accordance with one embodiment of the present disclosure;

[0022] FIG. 5 are images of an accelerometer used in connection with one embodiment of the voice enhancement system;

[0023] FIG. 6 is a printed circuit board of the voice enhancement system;

[0024] FIG. 7 is a portion of the voice enhancement system illustrating the positioning of the printed circuit board of FIG. 6 within a housing;

[0025] FIGS. 8-10 are electrical schematics forming the control arrangement of the voice enhancement system constructed in accordance with one example embodiment of the present disclosure;

[0026] FIG. 11 is a flowchart summarizing a method for increasing voice loudness in a patient in accordance with one embodiment of the present disclosure; and [0027] FIGS. 12-19 illustrate testing data from Parkinson's Disease patients and the results realized by the patients as a result of wearing the voice enhancement systems over various periods of time.

DETAILED DESCRIPTION

[0028] The present disclosure relates to a method and system for increasing voice loudness. More specifically, the present disclosure comprises an electrical voice enhancement system and method used to elicit the Lombard effect. The Lombard effect is an external cue for increasing voice loudness. The present method and system uses the Lombard effect to assist individuals experiencing problems with vocal intensity, articulation, and/or volume as a result of a physically degenerative condition, such as Parkinson's Disease. The Lombard effect produces an involuntary reaction in speakers to increase their voice loudness when speaking in noisy environments. In addition, the Lombard effect is known to not only influence the voice loudness in its speakers, but it can also alter the speaker's articulation quality, speech rate, and pitch.

[0029] Referring now to the figures, and in particular to FIG. 1, is a voice enhancement system 10 (hereinafter "VES" or "system") constructed in accordance with one embodiment of the present disclosure. The voice enhancement system 10 comprises a number of components and is designed to be used with patients diagnosed with physical ailments resulting in reduced vocal volume or intensity and experiencing problems with vocal loudness as a result of their disease. Parkinson's Disease, including idiopathic Parkinson's Disease is an example of one type of ailment treatable by the system 10; however, other diseases or speech disorders having similar ailments that can cause speech deficiencies, for example hypophonia, general articulation, low sound pressure level ("SPL"), high speech rates, reduced respiratory support, and poor vowel articulation are intended to be treated with the voice enhancement system 10 without departing from the spirit and scope of this disclosure. [0030] The voice enhancement system 10 as shown in FIG. 1 comprises a number of electrical components that are both internal and external to a housing 12. Located externally from the housing 12 is an activation device or accelerometer 14 and earpiece 16 that are coupled via feeds 18, 20 having connectors 22, 24, respectively to ports 26, 28 of the housing. In one example embodiment, the accelerometer 14 is an accelerometer manufactured by Knowles Acoustics of Itasca, IL. under part number BU-27135-000. In another example embodiment, the activation device 14 is a sensor capable of transforming energy from one form to another such as a transducer. In yet another example embodiment, the activation device 14 comprises a piezoelectric, piezoresistive, or capacitive type accelerometer.

[0031] The accelerometer 14 acts as an input device to the enhancement system 10.

The accelerometer 14 is relatively unaffected by noises in the environment. The accelerometer 14 was chosen, rather than a microphone, so that the enhancement system 10 would not be activated as a result of noise in the room or communication by a third person's speech. Stated another way, the activation device or accelerometer 14 provides an input signal to the voice enhancement system 10 that detects the initiation and duration of the patient's speech and is not activated by surrounding noise and/or non-patient noise.

[0032] During treatment, the accelerometer 14 is placed on any body part suitable for detection of initiation of patient's speech. In one example embodiment, illustrated in FIG. 2B the activation device 14 is worn on or attached to a portion of the patient's neck, such as on a skin surface adjacent one or both of the of thyroid lamina or in the sternal notch. In yet another example embodiment shown in FIG. 2A, the accelerometer 14 is worn on or attached to the surface of the patient's skin covering the temporal bone to receive bone conduction vibrations from the wearer to detect the onset of speech.

[0033] Examples of body parts suitable for detection of speech initiation using the system 10, in addition to the temporal bone and neck, include areas near the patient's mouth or lips. All of such body parts and positions are intended to be within the scope and spirit of the present disclosure.

[0034] In the example embodiment of FIG. 2A, the accelerometer 14 is attached in close proximity to the patient's ear. In both the embodiments of FIGS. 2A and 2B, the accelerometer 14 is attached to an epidermal surface using an adhesive.

[0035] The earpiece 16 acts as an output of the system 10, transmitting noise to the patient's ear during prescribed times during treatment. In one example embodiment, the prescribed time during a noise enabling condition starts when the patient initiates speech and continues while the patient talks and may continue for a prescribed duration when the patient ceases speech. The earpiece 16 is a mono-aural device that, in the illustrated example embodiment is non-occlusive to the patient's ear. The non-occlusive earpiece 16 advantageously allows the patient to hear their own speech during use of the system 10. Such advantageous results would not be experienced at the same level with an occlusive earpiece, which would have a tendency to obstruct the patient's hearing. An occlusive earpiece would have the effect of making the patient's voice sound louder to themselves, causing them to talk more quietly. Use of a non-occlusive earpiece avoids this negative effect.

[0036] The earpiece 16 having a support 19 is fed to the ear of the patient through thin tubing or feed 20 and an open ear fitting 30 as best seen in FIG. 2A. A suitable example of earpiece 16 that includes a feed 20 and an open ear fitting 30 is a product manufactured by Phonak AG of Switzerland under the name Fit'nGo Kit; however other open ear fittings made by other manufacturers could also be used with the system 10. The open ear fitting 30 is typically fit into the patient's ear by an audiologist. Use of an open ear fitting 30 avoids a reduction in vocal intensity due to the occlusion effect, which can occur with a closed ear fitting. [0037] In one example embodiment, the amplitude of the noise generated by the system 10 and transmitted to the output or earpiece 16 can be changed by a third party (e.g. a physician, speech-language pathologist, medical personnel etc.) treating the patient, but not by the user of the system. In the exemplary embodiment, the highest output level of the system 10 is less than 85 dBA, and ranges between 70 and 78 dBA, which is adequate to elicit the Lombard Effect and would not be expected to cause damage to the hearing mechanism or hearing of the patient that is being treated.

[0038] Referring now to FIG. 3 is a rear perspective view of the housing 12 used to support a portion of the electronics of the voice enhancement system 10. The voice enhancement system 10 illustrated in the example embodiment of FIGS. 1 and 3 is a portable system, allowing unrestrained mobility of the patient since it is designed to attach to the patient's body. In particular, rear view of the housing 12 shown in FIG. 3 illustrates a rear cover 32 of the housing having a support clip 34 integrated into the cover for attaching to the patient's clothing during use. Alternatively, the support clip 34 is used to hold the system 10 by attaching it to the patient's waist through either a belt clip or a fanny pack (depending on the patient's preference). The feeds 18, 20 can be fed underneath the patient's clothing to reduce the visible impact, that is, making the system 10 more inconspicuous to the patient's environment.

[0039] In the illustrated example embodiment of FIG. 1, the system 10 weighs no more than 6 ounces. Illustrated in FIG. 2C is an example embodiment of a voice enhancement system 10 comprising a compact design where the entire system 10 is positioned near the patient's ear. In particular, the non-occlusive earpiece 16, housing 12, and activation device (e.g. accelerometer) 14 are in contact with the patient's ear. In the example embodiment of FIG. 2C, the activation device 14 is cased partially within the housing 12 and is in contact with the surface of the patient's temporal bone and receives bone conduction vibrations from the patient to detect the onset of speech. The housing 12 in FIG. 2C contains all the electronics necessary, as further discussed below for processing signals from activation device 14 and producing noise 31 to the earpiece 16. The housing 12 also contains a power source such one or more batteries for supplying power to the system 10.

[0040] During treatment, the patient may wear the voice enhancement system 10 for several hours a day, increasing with treatment up to eight (8) hours per day. The treatment and design of the system 10 is such that it is highly mobile for the patient, allowing treatment to take place during daily living activities.

[0041] The system 10 is designed in such a way to externally cue the patient, for example, via the Lombard effect upon initiation of the patient' s speech, resulting in several positive and trained conditions in the patient, including increased sound pressure levels, reduced speech rate, improved respiratory support (reduced breathing pauses), and improved articulation. In one embodiment, the system 10 during treatment generates noise 31 that is projected from the earpiece 16 into the patient's ear upon the initiation of the patient's speech. In yet another example embodiment, the noise 31 is communicative unintelligible noise, simulating conversation between individuals, which is sent to the earpiece 16 worn in one of the patient's ears while he/she is talking. In one embodiment, the communicative noise 31 is generated from a product called Multitalker (20 Talkers) (MT) digital audio manufactured by AUDiTEC of St. Louis, MO. In yet another example embodiment, the noise 31 is white noise and/or random noise.

[0042] The presence of communicated noise 31 received by the earpiece 16 is an external cue to the patient to talk louder, naturally eliciting louder and clearer speech through the Lombard Effect. The Lombard Effect provoked by the use of the system 10 causes the patient to naturally and automatically speak louder under conditions of background noise generated by the system. The system 10 is believed to be most effective when the noise transmitted to the patient's earpiece 16 is more communicative in nature. However, random noise could also be transmitted to the patient's earpiece 16 without departing from the scope and spirit of this disclosure. Individuals that suffer from hypophonia, which is found in some forms of Parkinson's Disease can use the system 10 for treatment by wearing the system in natural communication contexts, achieving a louder, clearer, and more intelligible voice, without needing to self-cue.

[0043] Illustrated in FIG. 4 is a block diagram depicting a control arrangement 38 forming the voice enhancement system 10 in accordance with one example embodiment of the present disclosure. The control arrangement 38 in block diagram of FIG. 4 illustrates generally the components of the system 10, further shown in detail in FIGS. 8-10 and how the electrical components are interconnected. Centrally located within diagram is a central processing unit ("CPU") or microcontroller 40. In one embodiment, the microcontroller 40 is a 16 Bit 120K microprocessor. An example of a suitable microcontroller 40 is an MSP430F2618TPN manufactured by Texas Instruments.

[0044] The accelerometer 14 during use in one example embodiment is mounted on the patient's neck and is connected to electrical components 42 that are coupled the microprocessor 40 using feed 18 (see FIGS. 1 and 5). In one example embodiment, the feed 18 is a thin, flexible cable and a 3.5mm mini-phone audio connector 22 (see FIG. 1).

[0045] Referring again to FIG. 4, the connector 22 applies an output signal 43 upon speech by the patient generating movement detected by the accelerometer 14 to a preamplifier 44 with a gain of 2000. In one example embodiment, the accelerometer 14 activates the system 10 by sensing vibrations solely from the patient's vocal folds. An amplified signal 46 is transmitted through lead 48 that couples the preamplifier 44 to a bandpass filter 50. The band-pass filter 50 limits the frequency content of the signal 46 to form a filtered signal 52. In the illustrated example embodiment, the filtered signal 52 is limited by the band-pass filter 50 to a frequency content of approximately IOOHZ to 400HZ.

[0046] The filtered signal 52 is transmitted via lead 54 that couples the band-pass filter 50 to a comparator 56. The filtered signal 52 is then compared by the comparator 56 to a reference level 58 that is set by an adjustment 60 located on the housing 12. Every time the amplitude of the filtered signal 52 exceeds the reference level 58, the comparator 56 changes state, from low to high in an output signal 62. The output signal 62 is transferred to the microcontroller 40 via lead 64. In one example embodiment, the output signal 62 switches between 3.3 volts DC to 0 volts DC when changing from high to low state.

[0047] When the filtered signal 52 value drops back below the reference level 58, the comparator 56 changes state from high to low. This produces a stream of pulses in the output signal 62 that are applied to an interrupt 66 located within the microcontroller 40. As the state in the filtered signal 52 changes from high to low or vice versa, the noise transmitted from the system 10 into the earpiece 16 of the patient is enabled and disabled as the change in state occurs. The adjustment 60 that changes the reference level 58 allows medical personnel (such as physicians, nurses, speech-language pathologists etc.) treating the patient to manually optimize the sensitivity of the voice enhancement system 10 to the needs of each individual patient. Stated another way, the adjustment 60 allows the threshold for enabling and disabling the noise 31 received by the patient through the earpiece 16 to be adjusted based on the output signal 43 transmitted by the accelerometer 14.

[0048] The interrupt 66 of the microcontroller 40 uses a subroutine to analyze the pulses in the output signal 62 from the comparator 56 to determine when the patient begins speaking. Once it is determined that the patient is speaking, the microcontroller 40 begins reading the communicative noise 31 or audio 68, such as the product Multitalker (20 Talkers) (MT) digital audio manufactured by AUDiTEC of St. Louis, MO. from a micro memory card 70 that is coupled via lead 72 to the microcontroller. The microcontroller 40 then begins playing the communicative noise, random noise 31, or audio 68 through an amplifier 74 and a speaker 78 coupled to the microcontroller through lead 76 and feed 20 of the earpiece 16. In the illustrated exemplary embodiment, the amplifier 74 is a Class D amplifier and is combined with the speaker 78, using a digital to analog converter located within the microcontroller 40. The speaker 78 is connected to the patient's ear with thin clear plastic tubing of the earpiece 16. One suitable example of the earpiece 16 and speaker 78 is a product called Fit'nGo Kit open ear fitting manufactured by Phonak AG of Switzerland.

[0049] In one example embodiment, the microcontroller 40 is programmed 140 (see

FIG. 9) via software such that once the patient stops speaking for approximately 0.5 seconds, the microcontroller 40 stops playing the audio 68 or communicative noise 31. In another example embodiment, the software is programmed such that the audio or communicative noise 31 continues to occur for a range of approximately 500-750 milliseconds. This reduces breaks in audio 68 or communicative noise 31, which could be irritating to the patient during voiceless sound periods.

[0050] When the patient starts talking again, the microcontroller 40 continues playing the audio 68 or communicative noise 31 from where it stopped previously. In yet another example embodiment, the micro memory card 70 contains about 12 minutes and 30 seconds of communicative noise 31 or audio 68 data on a data file 80 located within the memory card. Once the entire audio data file 80 is played, the entire data file is started over at its beginning. This ensures that there is no obvious repetition of the audio generating the communicative noise 31 or audio 68.

[0051] The micro memory card 70 is also used to store data 82 about the usage of the voice enhancement system 10. When the audio or communicative noise 68 begins playing, a data record 84 is written to the memory card 70. Also when the audio 68 stops, another data record 86 is written to the memory card. The memory card 70 further contains a patient information record 88 that includes the patient number, as well as the date and time that the system 10 was initialized. Each patient information record 88 further contains audio 68 ON/OFF occurrences, elapsed time in days, hours, minutes, seconds and hundredths of seconds since the system 10 was initialized.

[0052] The voice enhancement system 10 is designed to be connected to a computer

90 using a serial interface 92. However, other interfaces 92, including USB, remote, and wireless connections for communicating the computer 90 and the system 10 are also viable forms of communication covered by the spirit and scope of this disclosure. The usage data 82-88 can then be downloaded from the system's memory card 70 via the interface 92 to the computer 90. In the exemplary embodiment, the usage data 82-88 is downloaded to the computer 90 using a program called PKTalker that is written in Lab VIEW.

[0053] The system 10 receives its power from a power supply 94. In the illustrated example embodiment, the source of the power supply 94 is two AA alkaline batteries that depending on usage, will provide power to the system to operate for approximately 7-10 days on one set of batteries. As best seen in FIG. 7, the batteries that act as the power supply 94 are located within the housing 12. In an alternative example embodiment (for example FIG. 2C), the power supply 94 supporting the system 10 is smaller and uses for example a rechargeable battery that is charged via a USB connection to the system.

[0054] Illustrated in FIG. 6 is a printed circuit board ("PCB") 96 used to support a portion of the electrical components 42 circuitry 100, microcontroller 40, and micro memory card 70 used within the system 10. The PCB 96 is located within the housing 12 as best seen in FIG. 7. The PCB 96 used by the system 10 is a four (4) layer PCB and constructed using surface mount components. [0055] In an alternative exemplary embodiment, the system 10 further comprises hardware to allow for external communication to a remote computer source. In one example embodiment, the system 10 includes a universal serial bus ("USB") or wireless connection, allowing communications with a remote computer for retrieving data and programming the data card 70.

[0056] In another alternative example embodiment, the system 10 is small enough for mounting to allow the accelerometer or transducer 14 to attach with a short connection to the patient's ear. One example embodiment is constructed such that the entire system 10 is small enough to be configured for positioning behind the patient's ear with the accelerometer 14 in contact with the patient's temporal bone (see FIG. 2A) to receive bone conduction vibrations from the wearer to detect the onset of speech.

[0057] In yet another alternative example embodiment, the system 10 uses an open wireless protocol, such as Bluetooth to deliver the audio to the patient's ear with a wireless connection to the accelerometer 14. In the alternative example embodiment, the system 10 is constructed to work with a Bluetooth headset, using processing capabilities of the microphone signal to determine when the patient is talking instead of the accelerometer 14.

[0058] The accelerometer 14 is coupled to the system 10 via the connector 22, which is shielded cable. In the illustrated example embodiment showing an electronic circuitry 100 of the system 10 in FIG. 8, the connector 22 is a 3.5mm mini-phone audio jack. Two ferrite beads, Ll and L2, are connected between an input lead 102 and reference lead 104 from the accelerometer 14 to an input 106 of the preamplifier 44 and an internal ground reference 108 of the system 10. The ferrite beads, Ll and L2 attenuate radio frequency noise picked up by the accelerometer 14 and connector 22. The ferrite beads Ll and L2 also attenuate radio frequency noise created by the system's microcontroller 40 to reduce the system's 10 radio frequency emissions. In the illustrated example embodiment, the size of the ferrite beads Ll and L2 have a 330 ohm impedance at 100 Mhz.

[0059] In FIG. 8, an electrostatic discharge suppressor ("ESD") 110 also identified in the electrical schematic as CRl protects the preamplifier 44 input 106 from static discharges. A resistor 112 also identified in the electrical schematic as R16 is an appropriate load resistor for the accelerometer. In the illustrated embodiment, the resistor 112 is a 5.6K Ohm .1W rated resistor. The dashed box representing the preamplifier 44 contains three (3) operational amplifiers ("OP-AMPS") that combine to produce a voltage gain of 2000X. A variable potentiometer 114 also identified in the electrical schematic as R3 is used to adjust the DC balance of the preamplifier 44. A test point 116 also identified in the electrical schematic as Jl is used to analyze the preamplifier 44 output and to adjust the DC level to 1.5 volts with no signal present.

[0060] The amplified signal 46 is transmitted by an output 118 from the preamplifier

44 along lead 48 to an input 120 of the band-pass filter 50. In the illustrated example embodiment, the band-pass filter 50 is a fourth (4^th) order band-pass filter that is centered at approximately 200HZ. A test point 120 also identified in the electrical schematic as J4 is used to observe an output 124 of the band-pass filter 50.

[0061] The output 124 of the band-pass filter 50 is connected to one input 126 of the comparator 56. Another input 128 of the comparator 56 is connected to the adjustable voltage reference 130 formed by R25, R27, R24 and C18. R27 is a potentiometer connected to a variable adjustment 60 that is accessible from the front panel 132 (see FIG. 1). The variable adjustment 60 in combination with potentiometer R27 is used to adjust the amplitude at which the system 10 detects that the patient is speaking through the vibration, movement of muscle, facial tissue, etc. , generated during speech. This allows the sensitivity of the system 10 to be set for a given patient. A test point 136 also identified in the electrical schematic as J3 is used to monitor the reference level during initial testing and adjustment. An additional test point, identified as reference character J6 in the electrical schematic is used to monitor the comparator output 62 to the microcontroller interrupt 66.

[0062] FIG. 9 shows that the output of the level comparator 58 (VOX) is connected to pin 37 (P4.1/TB1) of the TMS430F2618 microcontroller 40 also identified in the electrical schematic as U12. Pin 37 is configured in software to function as an interrupt. The system 10 includes a program 140 internal to the microcontroller 40 that uses this interrupt to trigger the playback of the audio 68 to the patient. The audio data 82-88 is stored in memory card 70 also identified in the electrical schematic as U15. The memory card 70 is connected to the microcontroller 40 using an SPI serial interface connection to UCBO (serial interface). The system program 140 also stores patient usage data 88 in flash memory 142 (see FIG. 9) every time the audio 68 is played.

[0063] Connector 144 also identified in the electrical schematic of FIG. 9 as J 12 is an

RS232 serial interface IC U16 used to connect the system 10 to a serial interface of a computer 90 so that the usage data 82-88 can be read from the flash memory 142 and the flash memory can be cleared of the usage data. The clearing of data in flash memory via software is understood by one skilled in the art. In the illustrated embodiment, the usage data 82-88 stored in the flash memory 142 is cleared using custom software written in Lab View. The software code or operating program 148 used to operate the system 10 and is downloaded into the microcontroller 40 using conventional interfaces appreciated by one skilled in the art. A JTAG connector 146 also identified in the electrical schematic as J13 is used during testing and loading of the system's 10 operating program 148 into the microcontroller's flash memory 142.

[0064] The amplifier 74 and speaker 78 module also identified in the electrical schematic as U9 is used to deliver the audio 68 to the patient. The amplifier 74 and speaker 78 are connected to the microcontroller 40 digital-to-analog output pin 5 also identified in the electrical schematic FIG. 9 as (DACO). In the exemplary illustrated embodiment, the audio 68 is delivered to the patient through a 0.05 inch ID 0.09 inch OD piece of TYGON® tubing 20 {see FIG. 1) manufactured by Saint-Gobain Performance Plastics Corporation of Aurora, Ohio. In the illustrated example embodiment, the audio data 80 is output from the microcontroller 40 at a sample rate of 8kHZ.

[0065] Resistor 150 and capacitor 152, also identified in the electrical schematic as

R31 and C30, respectively form a low pass filter 154 used to attenuate converter artifact. In the illustrated example embodiment, the low pass filter 154 attenuates converter artifact that is over approximately 5kHZ. The externally accessible potentiometer 134 also identified in the electrical schematic as R28 is used to adjust the amplitude of the output audio 68 signal. A test point 156 also identified in the electrical schematic as J9 is used to measure the audio output signal 68 during device testing.

[0066] FIG. 10 illustrates the schematics for the device power supplies 94. A screw terminal connector 158 also identified in the electrical schematic as JlO is coupled to the two battery power source. In the illustrated embodiment, two (2) AA batteries are used to power the system 10. In an alternative example embodiment, the power supplies 94 comprise a single or multiple rechargeable battery or batteries. L3 and L4 are ferrite beads used to attenuate RF interference. In the illustrated embodiment, the L3 and L4 ferrite beads have a 330 ohm impedance at 100 Mhz.

[0067] A schottky diode 160 also identified in the electrical schematic as CR2 is used to protect the power supply 94 from backward connected batteries. A switched capacitor power supply regulator IC 162 also identified in the electrical schematic as U 14 produces a main DC power supply 164 for the system 10. In the illustrated embodiment, the main DC power supply 164 provides 3.3 volts DC of power to the digital circuitry of the system 10. A test point 166 also identified in the electrical schematic as J5 connects a reference for measurements made during device setup and testing. A test point 168 also identified in the electrical schematic as JIl is used to check the main DC power 164.

[0068] A low dropout linear regulator 170 also identified in the electrical schematic as

UI l provides regulated DC power for the analog and audio circuitry in the system 10. In the illustrated example embodiment, the linear regulator 170 provides 3.0V DC power to the analog and audio circuitry in the system 10. A test point 172 also identified in the electrical schematic as J8 is used to check the 3.0V DC supply. A low dropout linear regulator 174 also identified in the electrical schematic as UlO is used to provide 1.5 volt DC power for the system 10. The regulator 174 is used to power the amplifier 74 and speaker 78 as well as for a pseudo reference for the OP-AMPS in the preamplifier 44 and filter 50. A test point 176 also identified in the electrical schematic as J7 is used to check the 1.5V DC supply.

[0069] Illustrated in FIG. 11 is a flowchart summarizing a method 200 for increasing voice loudness in a patient in accordance with one embodiment of the present disclosure. At 210, the method 200 comprises positioning an accelerometer on a patient. At 220, the method 200 comprises transmitting an output signal from the accelerometer when the patient is speaking. At 230, the method 200 comprises analyzing the output signal against a prescribed threshold. At 240, the method 200 comprises transmitting audio noise to the patient from a microcontroller when the output signal is above the threshold.

[0070] An additional benefit of the system 10 is that the patient will be trained to use a louder voice even when not wearing the system over the course of the treatment period, leading to an extended therapeutic effect. For example, patients after using the system 10 for an extended period of time will produce louder and clearer speech, increasing a number of decibels (dB) in SPL than experienced at the start of treatment without the device on. This therapeutic effect will beneficially grow over a treatment period using the system 10 on and off the patient, allowing the patient to maintain louder and clearer speech between longer treatment periods. Testing Results and Training of Parkinson's Disease Patents ("PDPs")

[0071] FIGS. 12-19 and discussion below involve testing results from training PDPs using or having the voice enhancement system 10 "on" and "off over an eight (8) week period. The PDPs realized a number of positive changes in their communication as a result of the training and use of the system 10. The positive changes that were realized discussed further below include: increased Sound Pressure Levels; reduced Speech Rate; improved breath support; and improved Vowel Articulation.

[0072] FIGS. 12-19 relating to the testing results illustrate data taken from one or two sessions before the system 10 was positioned on the patient or user is labeled "pre". The data taken four (4) weeks into the eight (8) week training period (when available) is labeled "Tr2". And, the data taken as the end of the eight (8) week training period is labeled "Pol". At all data points shown in FIGS. 12-19, the data was measured with the system 10 off (labeled "Of) and then on (labeled "On"). When the patient was "on" or using the system 10, the training consisted of the patients wearing the system in communicative environments for 4-6 hours per day for eight (8) weeks. The PDPs returned every two (2) weeks for evaluation. Testing then occurred with the patients having the system 10 off and on.

Sound Pressure Level

[0073] Sound Pressure Level ("SPL") is a measure of the intensity of the voice.

PDPs often have weak, quiet voices, making vocal intensity a major therapy target. Referring now to FIG. 12, the SPLs (shown along the vertical axis) were higher with the system 10 on than when the system 10 was off due to the Lombard Effect. The SPL data in FIG. 1 was collected from an extemporaneous speech task where PDPs talked about a topic of their choice for two (2) minutes. This task is indicative of real- world speech production. In FIG. 1, it is clear that SPL increased when the system 10 was on. This increase in SPL levels lasted throughout the eight (8) week training period. The participants charted in the testing illustrated in FIG. 1, resulted in no maintenance of the SPL effect without the use of the system 10.

Speech Rate

[0074] Speech Rate is a measure of the number of syllables produced per second.

PDPs sometimes speak more quickly than typical speakers, making a reduction in rate a therapy target. When typical adults speak more loudly (including under the Lombard effect), they have been shown, in some studies, to slow their speech rate.

[0075] The effect of loudness on speech rate can depend on what the person is saying.

Speech rate data was collected (see FIG. 13) in the study from an extemporaneous speech task where patients talked about a topic of their choice for two (2) minutes. This task is indicative of real- world speech production. It is clear when look at FIG. 13 that when the system 10 is first used, the speech rate declines markedly. After four (4) weeks of training, the patients' speech rate off use of the system 10 is lower than at baseline, but the system 10 results in an increase in speech rate. By the end of the eight (8) weeks of training, the patients' speech rate off or during non-use of the system 10 is still lower than at baseline, and the system results in a small decrease in speech rate. Such results suggest that the system 10 provides a benefit to speech rate, even when the system 10 is not on after eight (8) weeks of training and therapy. A slower speech rate is particularly important helping listeners to understand what the patient is saying.

Breath Support

[0076] Breath Pausing behaviors can provide information about the patient's respiratory support for speech. PDPs have reduced respiratory support and have been reported to not use pauses, except when they need to breathe. Typically adults pause both to breathe (breath pauses) and for emphasis (non-breath pauses). Evidence that the system 10 trains respiratory support would result in an increase non-breath pauses and a decrease in breath pauses.

[0077] During the study, data was collected relating to the frequency of the breath pausing through an extemporaneous speech task where patients talked about a topic of their choice for two (2) minutes. FIG. 14 depicts the percent of breath pauses (out of the total number of pauses). When the system 10 is first put on the patient, the number of breath pauses increase. To talk louder, the patients must use more respiratory muscle effort and increase the number of breaths taken. PDPs may find that, in the beginning of treatment, they need to take more breaths to support speech when the system 10 is on. However, after eight (8) weeks of therapy and training, the percent of breath pauses as illustrated in FIG. 14 declined. As a result, the system 10 trains and conditions PDPs to take fewer breath pauses and more non-breath pauses. Advantageously, the system 10 after training the PDP for a period eight (8) weeks speaks more loudly and does not require the patient to take more breath pauses. These data suggest that the system 10 improves respiratory support for speech.

Vowel Articulation

[0078] Vowel Articulation can be indexed by looking at the resonances produced in the vocal tract for point vowels. The three point vowels in English are "ee" as in "feet", "oo" and in "boot", and "ah" as in "dot". The resonances are called the first and second formant frequency. As vowel articulation improves, the distance between the formants for the point vowels will increase. PDPs often have slurred or indistinct articulation. Some studies have shown that when typical adults and PDPs increase SPL, articulation becomes clearer. Improving articulation involves making the acoustic distance between the vowels larger (relative to the first and second formants), suggesting that they are perceptually more distinct as well. Evidence that the system 10 trains PDPs relating to vowel articulation would result in the vowels becoming more distinct when the patients are using the system due to the Lombard effect. The data in FIG. 15 was collected during the study from a sentence repetition to ensure that the analysis included the same words for each vowel across patients.

[0079] FIG. 15 depicts the formant frequencies for a female patient. For the vowel

"oo" (lower left corner), the first formant frequency decreased, in the direction for improved articulation for the vowel "oo". For the vowel "ee" (upper left corner), the first formant frequency decreased and the second formant frequency increased. Both of these changes are in the direction for improved articulation for the vowel "ee".

[0080] The results of the study show more positive effects due to training occurs from the use of the system 10 rather than the improvements realized by just using the system. When the system 10 is first used, there is little change except for the vowel "ee". The positive effects of the system 10 after eight (8) weeks of training do not depend on the system being on. Data for the "ee" and "oo" vowels after training (Pol) off and on the system 10 are similar, suggesting improved articulation is learned, not just present when the system 10 is on.

[0081] Finally, the depicted effects do not depend strictly on increased intensity of speech. For both "ee" and "ah", louder speech should result in a larger mouth opening and higher first formant frequencies. For "ee", this would not improve vowel clarity. The end result of treatment is a lower first formant frequency.

[0082] FIG. 16 depicts the formant frequencies for a male patient. For the vowel

"oo" (lower left corner), the first and second formant frequencies decreased, in the direction for improved articulation for the vowel "oo". For the vowel "ee" (upper left corner), the second formant frequency increased, but only in the post-treatment off condition, in the direction for improved articulation for the vowel "ee". For the vowel "ah", the first and second formant frequencies decreased, in the direction for improved articulation for the vowel "ah".

[0083] The results of the study show more effects due to training than the positive effects that are realized by just putting the system 10 on a PDP. When the system 10 is first used, there is no change in the formants. The effects of using the system 10 after eight (8) weeks of training do not depend on the system being on. Data for the "oo" vowel after four

(4) and eight (8) weeks of training (Tr2 and Pol) off and on the system 10 are distinct from the pre-data, suggesting improved articulation is learned, not just present when the system 10 is on.

[0084] Finally, the depicted effects do not depend strictly on increased intensity of speech. For both "ee" and "ah", louder speech should result in a larger mouth opening and higher first formant frequencies. For "ee", this would not improve vowel clarity. The end result of treatment from the system 10 is a lower first formant frequency for both "ee" and

"ah".

Long- Term Testing Results and Training of PDPs

[0085] FIGS. 17-19 illustrate long-term data relating to two PDP subjects. The methodology, periods (pre-training, training, and post-training), and nomenclature ("pre" "Tr2"

"Pol") is similar to that for the patients described above. Also similar in the methodology of the long-term testing was when the patient was "on" or using the system 10, the training consisted of the two subjects wearing the system 10 in communicative environments for 4-6 hours per day for eight (8) weeks. The PDPs subjects then returned every two (2) weeks for evaluation. Testing then occurred with the two subjects having the system 10 on and off.

First Subject

[0086] Below are data from the first PDP subject who used the system 10 for the regular training period (eight (8) weeks) and then was "off; that is, the first subject did not use the system 10 at all for four (4) weeks (referred to as the "detraining" period). The first subject then used to use the system 10 again (after the four (4) off period) and the subject's progress was tracked for another six (6) months while the subject was using or on the system 10. The first subject was frequently tested without the system 10 on (that is, not wearing the system 10 during testing) to check for learning/training effects. However, three (3) and six (6) months into the final six (6) month period with the system 10 on, the first subject was tested first without the system 10 on (i.e. not wearing the system 10 during testing) and then with the system on (i.e. wearing the system 10 during testing).

[0087] Looking at the sound pressure level (SPL) data in FIG. 17, it is clear that:

1. The ability of the system 10 to affect SPL is maintained for the entire treatment period (a total of nine (9) months - two (2) during initial training, and six (6) during training again).

2. While the first subject does not show a training effect of increased SPL off the system 10, during the eight (8) week training period the subject does show a training effect off the system starting at month three (3), suggesting the first subject needed longer than eight (8) weeks to train to a higher SPL.

[0088] Illustrated in FIG. 18 is the first subject's speech rate. The first subject's speech rate is lower off (i.e. not wearing the system 10 during testing) the system 10 starting after four (4) weeks of training. The training effect is maintained (speech rate lower than pre values) for the entire ten (10) months.

Second Subject

[0089] Illustrated FIG. 19 are SPL data from a second subject who was on the system

10 for the regular training period (eight (8) weeks) and then was off the system 10 for five (5) months. The second shows an increase in SPL off (i.e. not wearing the system during testing) the system 10 after four (4) weeks of training, and that training effect is higher at the end of treatment (Post). SPL is variable in the five (5) month off period, but for all four testing periods, SPL is higher than data taken from two sessions before the system 10 was positioned on the patient or user is labeled "pre".

[0090] To summarize, FIGS. 12-19 along with the discussion above demonstrate that the

PDP subjects realize improved communication characteristics throughout training and while wearing the system 10. These improved communication characteristics include increased sound pressure levels, eliciting louder speech, reduced speech rate, improved respiratory support (reduced breathing pauses), and improved vowel articulation. The data discussed in connection with FIGS. 12-19 indicate that the improved communication characteristic of eliciting louder speech continues over a period of at least six (6) months while the system 10 is on (i.e. the patient is wearing the system).

[0091] The test results also indicate that the improved communication characteristics of at least sound pressure level ("SPL") and speech rate continue when the system 10 is off and not worn at all by the patient over at least a one-month period. This provides evidence that the system 10 trains and conditions the patient. Stated another way, the system 10 creates improved communication characteristics not only throughout training and while the system is being worn by the patient, but also after the patient ceases wearing or using the system.

[0092] What have been described above are examples of the present invention. It is, of course, not possible to describe every conceivable combination of components or methodologies for purposes of describing the present invention, but one of ordinary skill in the art will recognize that many further combinations and permutations of the present invention are possible. Accordingly, the present invention is intended to embrace all such alterations, modifications, and variations that fall within the spirit and scope of the aforementioned disclosure.

Claims

In the Claims:

1. A system to improve communication comprising:

an assembly fixedly positioned during use in close proximity to a user's ear; said assembly comprising an accelerometer to detect the initiation and duration of the user's speech and an output presentation system;

said output presentation system comprising a non-occlusive ear fitting that presents noise that is unrelated to the sound- frequency or intonation of the user's speech; and a control arrangement to maintain presentation of said noise throughout the detected duration of the user's speech.

2. The system to improve communication of claim 1 wherein the user has been diagnosed with Parkinson's Disease.

3. The system to improve communication of claim 1 wherein said noise is one of white noise, random noise, and unintelligible noise.

4. The system to improve communication of claim 1 wherein said control arrangement further comprises a comparator having a prescribed threshold such that movement to said accelerometer resulting in an output above said threshold generates said noise until said output drops below said threshold.

5. The system to improve communication of claim 4 wherein said noise continues for a prescribed period of time after said output drops below said threshold.

6. The system to improve communication of claim 1 wherein the user of the system experiences an improved communication characteristic comprising sound pressure level or speech rate and that said improved communication characteristic continues following the removal of system from the user for a period of at least one month.

7. A method to improve communication in a patient with Parkinson's Disease comprising:

training the patient' s speech production mechanism and/or speech motor control mechanisms through a voice enhancement system that introduces noise to the patient during the patient' s speech.

8. The method to improve communication in a patient with Parkinson's Disease of claim 7 wherein the noise introduction is achieved by using an activation device that senses body movement generated or vibration by the patient during speech and upon sensing said movement or vibration the activation device responds by producing an output signal during the patient' s speech to the voice enhancement system resulting in the introducing the noise to the patient.

9. The method to improve communication in a patient with Parkinson's Disease of claim 8 further comprising providing a control arrangement having a prescribed threshold such that output signals generated by the activation device above the threshold result in the introduction and continued generation of noise to the patient.

10. The method to improve communication in a patient with Parkinson's Disease of claim 9 further comprising communicating the output signals from the activation device to the control arrangement wirelessly.

11. The method to improve communication in a patient with Parkinson's Disease of claim 7 further comprising removing the voice enhancement system from the patient; wherein the patient continues to have improved communication in the form of sound pressure level or speech rate following the removal of the voice enhancement system for a period of at least one month.

12. A system for training a patient with Parkinson's Disease to have improved communication, said system comprising;

an activation device that is responsive to movement or vibrations in a patient during speech and upon sensing the patient's speech movement or vibration, produces an output signal;

a presentation assembly comprising an ear fitting that transmits noise to the ear of a patient during a noise enabling condition; and

a control arrangement coupled to said activation device for receiving said output signal, said control arrangement comprising a signal processor having a prescribed threshold that is analyzed against said output signal such that one or more output signals above said threshold generates said noise enabling condition resulting in noise transmission from said control arrangement to said presentation assembly.

13. The system of claim 12 wherein at least a portion of said activation device is externally located on a patient's epidermis.

14. The system of claim 12 wherein said noise enabling condition continues only while said output signal remains above said prescribed threshold.

15. The system of claim 12 wherein said noise enabling condition continues for a prescribed duration once the output signal drops below said prescribed threshold.

16. The system of claim 12 wherein said signal processor further comprises an amplifier, band-pass filter, and comparator.

17. The system of claim 12 wherein said control arrangement further comprises a microcontroller having memory for communicating said noise to said presentation assembly during said noise enabling condition.

18. The system of claim 12 wherein said activation device is one of an accelerometer and transducer.

19. The system of claim 12 wherein said noise is white noise, non-communicative noise, and/or random noise.

20. The system of claim 17 wherein said memory further records frequency and duration of the noise enabling conditions experienced by a patient during use of said system.

21. The system of claim 12 wherein said ear fitting further comprises a non- occlusive ear fitting.

22. The system of claim 15 wherein said prescribed duration for the continuation of noise enabling condition once the output signal drops below said prescribed threshold occurs for a time period ranging between 500 and 750 milliseconds.

23. A method for training a patient to have at least one improved communication characteristic, the method comprising:

positioning a non-occlusive ear fitting in a patient's ear that transmits noise to the ear of the patient during a noise enabling condition;

positioning externally an accelerometer on the patient at a location that moves or vibrates during speech;

transmitting an output signal from the accelerometer to a control arrangement upon sensing movement or vibration experienced during speech of the patient;

analyzing the output signal with a signal processor within the control arrangement against a prescribed threshold; and

transmitting audio noise to the ear fitting from the control arrangement when the output signal is above the threshold, generating said noise enabling condition.

24. The method of claim 23 further comprising positioning said control arrangement on the patient.

25. The method of claim 24 further comprising positioning said control arrangement in contact with the patient' s ear.

26. The method of claim 23 further comprising:

removing the non-occlusive ear fitting and accelerometer from the patient; wherein the patient continues at least one improved communication characteristic following the removal of the non-occlusive ear fitting and accelerometer.

27. The method of claim 26 wherein said at least one improved communication characteristics are sound pressure level and speech rate.

28. The method of claim 27 wherein said at least one improved communication characteristic comprising sound pressure level and speech rate continue for a one-month period following the removal of the non-occlusive ear fitting and accelerometer.

29. The method of claim 23 wherein said at least one improved communication characteristic comprises one of increased sound pressure level, reduced speech rate, improved breath support, and improved articulation.