WO2018134558A1

WO2018134558A1 - An electronic fluency device

Info

Publication number: WO2018134558A1
Application number: PCT/GB2018/000013
Authority: WO
Inventors: Dimitri VAN IMMERSEEL; James Bird; Jordi FERNANDEZ OROMI; Rowan BRACKSTON; Juan Marcos GARCIA DE LA CRUZ; Robert SIDDALL
Original assignee: Respira Limited
Priority date: 2017-01-20
Filing date: 2018-01-22
Publication date: 2018-07-26
Also published as: GB201701052D0

Abstract

An electronic fluency device (1) comprises a sensing means (2, 3) configured to monitor the breathing of a user wearing the device (1), and the speech patterns of a user; a processing means (7) configured to receive inputs from the sensing means (2, 3) relating to the breathing and speech patterns of a user, and to assess the inputs against baseline speech and breathing patterns, and; a warning means (6) configured to provide an alert to a user, the warning means (6) configured to receive a control signal from the processing means (7) to initiate the alert, and wherein; the processing means (7) is further configured to transmit the control signal if the breathing and/or speech patterns of a user fall outside the baseline speech and breathing patterns.

Description

An electronic fluency device

FIELD

The present invention relates to an electronic fluency device for assisting with speech. More particularly, the present invention relates to a device for monitoring biometric data and speech patterns for assisting with speech therapy. Even more particularly, the present invention relates to a device that provides real-time monitoring of the biometric data and speech patterns of a user in order to provide real-time feedback and/or performance history tracking for encouraging correct breathing, and/or speaking habits to improve fluency, The present invention also relates to a method of monitoring biometric data and speech patterns for assisting with speech therapy.

BACKGROUND

It is not uncommon for individuals to have disordered speech patterns such as stuttering or similar. Therapy for speech disorders such as stuttering is intended to assist a person by helping them to form different breathing and speech habits that assist with their speech, and to help them to avoid speech and breathing patterns or habits that are known to be potentially problematic. There are many proven psychological and cognitive practices that have been shown to help people who stutter (PWS) gain fluency. Some of these practices include using the full capacity of the diagram when breathing, limiting the number of words spoken per breath and per minute, using a low pitch of voice and acceptance of oneself as a PWS.

Therapy for speech disorders such as stuttering is intended to assist a person by helping them to form different breathing and speech habits that assist with their speech, and to help them to avoid speech and breathing patterns that are known to be potentially problematic. However, people who stutter tend to use or 'activate' their techniques only when they are in difficult speaking situations, outside of their comfort zone. These techniques can be ineffective if only used when fear is high.

A number of devices are known that provide monitoring of speech patterns for speech therapy.

US4,020,567 describes and shows a method and apparatus for the detection of certain characteristics in the speech of stutterers as they participate in a program which reconstructs the basic properties of speed sounds. The method of speech therapy for stutterers includes transducing a patient's speech into an electrical signal and comparing the transduced signal to a reference signal representative of the desired speech target behaviour. The comparison is initiated when the amplitude of the transduced speech signal exceeds a threshold related to minimal sound level. When the amplitude of the transduced speech signal exceeds that of the reference signal, the subject is visually advised of an error in his speech. When the amplitude of the transduced speech signal remains above the trigger threshold and below the reference signal, the subject is visually advised that a correct speech response has been made. In a second embodiment, the rate of change of the amplitude of the subject's speech signal is compared with a reference related to the desired target behaviour.

US5940798 describes and shows a treatment system for reducing stuttering that uses an auditory feedback modification technique to train the stutterer's speech motor control system to be more stable. The auditory feedback modification is based on a model of speech motor control in the human brain that incorporates a variation of observer-based control and Smith prediction. In addition, the Kalman gain of the model is set by comparing the speech muscular control signals sent to the person's vocal tract to the corresponding auditory speech sounds the person actually hears. It is believed that the speech motor control system of a stutterer has set the Kalman gain too high, thereby creating an unstable control system that in turn causes stuttering. A feedback modifier feeds back the stutterer's speech to the stutterer with a perturbation that is small enough to pass a validator function that is believed to be part of the speech motor control system. The small perturbations increase the difference between the target speech and the fed back speech, which is believed to cause the speech motor control system to decrease the Kalman gain. Over a treatment program, it is believed that the perturbations "train" or adapt the stutterer's speech motor control system so as to reduce the Kalman gain, which is further believed to persist when the auditory feedback is discontinued.

In this specification where reference has been made to patent specifications, other external documents, or other sources of information, this is generally for the purpose of providing a context for discussing the features of the invention. Unless specifically stated otherwise, reference to such external documents is not to be construed as an admission that such documents, or such sources of information, in any jurisdiction, are prior art, or form part of the common general knowledge in the art. SUMMARY

It is an object of the present invention to provide an electronic fluency device for assisting with speech which goes some way to overcoming the abovementioned disadvantages or which at least provides the public or industry with a useful choice. it is a further object of the invention to provide a method of monitoring biometric data and speech patterns for assisting with speech therapy which goes some way to overcoming the abovementioned disadvantages or which at least provides the public or industry with a useful choice.

The term "comprising'' as used in this specification and indicative independent claims means "consisting at least in part of. When interpreting each statement in this specification and indicative independent claims that includes the term "comprising", features other than that or those prefaced by the term may also be present. Related terms such as "comprise" and "comprises" are to be interpreted in the same manner.

As used herein the term "and/or" means "and" or "or", or both.

As used herein "(s)° following a noun means the plural and/or singular forms of the noun.

Accordingly, in a first aspect the present invention may broadly be said to consist in an electronic fluency device, comprising: at least one sensor configured to measure biometric data relating to the speech of a user; a processor configured to receive input from the at least one sensor, and to assess the input against baseline speech and breathing patterns; a warning mechanism configured to provide an alert to a user, the warning mechanism configured to receive a control signal from the processor to initiate the alert; the processor further configured to transmit the control signal if the biometric data falls outside baseline values.

In an embodiment in use, the at least one sensor is configured to be worn by a user.

In an embodiment, the electronic fluency device further comprises a chest belt configured to extend around the chest of a user.

In an embodiment, the electronic fluency device further comprises a box, the chest belt connected to and extending from the box, at least one of the sensor or sensors located in the box.

In an embodiment, at least one of the sensor or sensors is configured to locate on the belt. In an embodiment, the sensor or sensors is/are configured to measure costal rib expansion and contraction.

In an embodiment, the sensor or sensors comprises a strain gauge/load cell.

In an embodiment, the sensor or sensors comprises a chest impedance and heart rate sensor configured to assess chest impedance.

In an embodiment the sensor or sensors comprises at least one microphone.

In an embodiment, the at least one microphone comprises at least one inwards- facing microphone.

In an embodiment, the at least one microphone comprises at least one outwards- facing microphone.

In an embodiment, the sensor or sensors comprises an accelerometer.

In an embodiment, the warning mechanism comprises a haptic actuator transducer configured to provide a vibration against the chest of a user when activated.

In an embodiment the processor comprises a programmable device.

In an embodiment, the processor comprises a microcontroller.

In an embodiment, the programmable device is integral with the sensor.

In an embodiment, the programmable device comprises a remotely located programmable device, in communication with the sensor via wireless communication.

In an embodiment, the remotely-located programmable device is a mobile device. In an embodiment, the programmable device is configured to receive readings from the sensor or sensors and compare these to preset baseline readings.

In an embodiment, the programmable device is configured so that the preset baseline readings can be adjusted.

In an embodiment, the processor is further configured to store and/or transmit data relating to one or both of breathing and speech patterns.

In an embodiment the control signal comprises multiple signals, each signal causing the warning mechanism to provide a different alert, the different alerts indicative of a particular pattern of speech or breathing falling outside baseline speech and breathing patterns. in a second aspect the present invention may broadly be said to consist in a method of monitoring biometric data and speech patterns for assisting with speech therapy, comprising the steps of:

i) measuring biometric data relating to the speech of a user;

ii) comparing the measured biometric data against baseline speech and breathing patterns;

iii) providing an alert to a user if the biometric data falls outside the baseline values.

In an embodiment, in the step of measuring biometric data, the biometric data consists of one or more of: costal rib expansion and contraction; chest impedance; the voice of the user; ambient external background noise; heart rate.

In an embodiment in the step of comparing measured biometric data against baseline speech and breathing patterns, the user's voice is monitored, and patterns for the user are identified, the alert provided if the patterns move towards a disordered speech pattern.

In an embodiment, the user's voice is specifically monitored for an increasing number of words between breaths, and/or an increasing frequency of words for a set time period.

In an embodiment, the user's voice is monitored using a user-specific microphone, and ambient noise is monitored using a separate microphone, the difference between the two used to isolate the characteristics of the user's voice.

With respect to the above description then, it is to be realised that the optimum dimensional relationships for the parts of the invention, to include variations in size, materials, shape, form, function and manner of operation, assembly and use, are deemed readily apparent and obvious to one skilled in the art, and all equivalent relationships to those illustrated in the drawings and described in the specification are intended to be encompassed by the present invention.

This invention may also be said broadfy to consist in the parts, elements and features referred to or indicated in the specification of the application, individually or collectively, and any or all combinations of any two or more said parts, elements or features, and where specific integers are mentioned herein which have known equivalents in the art to which this invention relates, such known equivalents are deemed to be incorporated herein as if individually set forth. Therefore, the foregoing is considered as illustrative only of the principles of the invention. Further, since numerous modifications and changes will readily occur to those skilled in the art, it is not desired to limit the invention to the exact construction and operation shown and described, and accordingly, all suitable modifications and equivalents may be resorted to, falling within the scope of the invention.

BRIEF DESCRIPTION OF THE FIGURES

Further aspects of the invention will become apparent from the following description which is given by way of example only and with reference to the accompanying drawings which show an embodiment of the device by way of example, and in which: Figure 1 shows an embodiment of the electronic fluency device of the present invention in use, worn on the chest of a user, the device comprising a chest belt configured to pass around the chest to allow the user to wear the device and to hold the device in place, and an enclosed central box that contains sensing and feedback elements.

Figure 2 shows detail of the enclosed central box of figure 1 , the box containing inwards and outwards facing microphones, a chest impedance and heart rate sensor, a haptic actuator transducer, a processor and a wireless communication module, a strain gauge and load cell.

Figure 3 shows a variation of the device of figures 1 and 2, the device in use and worn on the chest of a user, the device of this variation having chest impedance and heart rate sensors located on the belt each side of the central box, and an

accelerometer located within the central box as an additional biometric sensor.

DETAILED DESCRIPTION

An embodiment of the electronic fluency device of the present invention will now be described with reference to the figures.

An embodiment of the device is shown in figure 1.

Structure

The device 1 of the embodiment shown has a chest belt 3, which in use extends and connects around the chest of a user in a similar manner to a sports heart rate monitor. The chest belt 3 is made from elasticated fabric or similar, and has a clasp at the rear that allows a user to easily connect and disconnect the free ends to wear or remove the device 1. An enclosed box 2 is attached to the belt at the centre of the belt 3, so that when the device 1 is worn, the box 2 is located at the front centre of the user's chest, at or substantially at the same position or height as the diaphragm of a user, the box 2 located over or just below the base of their sternum. The box 2 contains a number of sensing and feedback elements as detailed below, which in use are intended to measure and monitor various biometric data and the speech patterns of a user. The position of the belt 3 when worn around the chest ensures that the sensors are particularly sensitive to 'costal breathing'. This helps to ensure that the sensors provide useful feedback and sensing for people who stutter, since a greater expansion of the ribs means a greater inhalation using the full diaphragm. The sensors are connected to the belt, via their connection to and location inside the box. As outlined elsewhere, all of the sensors can be located within the box, or certain sensors can be located on the belt itself in variations of the main embodiment.

The box 2 has a smoothly contoured outer surface, so that it is low-profile under the clothes of a user, and is comfortable when the device 1 is wom. In this embodiment, the belt is divided into two separate halves, each half extending from opposed sides of the box, so that the inner surface of the box can sit flush against the skin of a wearer/user.

A number of sensing and control elements are present in the box 2, as follows: an inwards-facing microphone or microphones 4a, an outwards-facing microphone or microphones 4b, a chest impedance and heart rate sensor 5, a haptic

actuator transducer 6, a processor and a wireless communication module 7, a strain gauge and load cell 8, and an accelerometer 9. A battery (not shown) is also present in the box 2, and is used to power those elements of the device 1 which require it. In variations of this embodiment, the strain gauge and load cell 8, and the chest impedance and heart rate sensor 5 could replace one another (i.e. each could be used singly, without the other present). In this embodiment the wireless

communication module 7 comprises a wireless transmitter/receiver. This could for example be a Bluetooth module. However, this module could use any suitable wired or wireless device to make an external connection.

The function of these elements is to provide biometric data relating to the speech of a user, as follows:

The microphone(s) 4 monitor the sound waves created by the lungs during breathing and speech by a user. The microphones are intended mainly to track the speech patterns of the user. Microphone 4a listening inwards can be used in combination with the strain gauge sensor 8 and the chest impedance sensor 5 to improve the robustness of the breathing measurements.

The chest impedance and heart rate sensor 5 has electrodes, which are configured to make electrical connections between the circuitry from the box 2 and the skin of the user, in order to allow the heart rate and the inductance of the chest to be determined. In this embodiment, the electrodes are located both on the belt 3 and the inner surface of the box 2. The electrodes terminate in contact with the skin of the user at various locations along the belt 3 in use (and the inner surface of the box 2). Using a chest impedance sensor offers an alternative to the use of a strain gauge or gauges, as it is a solid-state component that can be used to determine the frequency and depth of a user's breath, and the readings can be used in parallel with similar readings from the strain gauge/load cell 8 (see below). Data from one of the chest impedance and heart rate sensor 5 or the strain gauge/load cell 8 can be used in making an assessment, or the data can be combined. The chest impedance and heart rate sensor 5 offers a solid-state alternative to the strain gauge(s) 8 for determining the frequency and depth of the user's breath, although it is used generally for the same purpose. In this embodiment, a combination of the data of both is used to increase the performance of the breath measurement, based on machine learning algorithms.

The data from the chest impedance and heart rate sensor 5 will give an insight into the mental and psychological state of the user, for example indicating if the user is more likely to stammer in high stress situations. In this embodiment, the data is processed or analysed in the processor and shared via the communication module to a remote app or another remote computing device as required, where further processing can occur. The measured and recorded heart rate can also help to measure breathing rate.

The haptic actuator/transducer 6 is essentially a vibrating unit or vibrator, that is configured to provide a vibration against the chest of a user when certain parameters are met or not met This provides a physical alert or alarm to the user, that allows them to change or reset their current speech or breathing pattern to something more preferred. The haptic actuator/transducer 6 is capable of producing different patterns of vibration or tones in order to provide different messages in relation to breathing depth, speech speed, etc.

The strain and load gauge 8 monitors the movement of a user's chest as they breath and talk. The strain gauge/load cell 8 measures the deformation of the chest cavity, which is a measure that relates to the chest impedance. These readings are used to infer the depth and rate of breath of the user. These reading can be used in conjunction with the readings from the chest impedance and heart rate sensor 5, or separately to these.

The accelerometer 9 is used to measure the movement of the user's chest as they breathe and talk - acceleration/motion data. Multiple accelerometers can be used if required. The accelerometer(s) 9, which in this embodiment is/are located within the box 2, but which may be embedded within the belt 3 in variations, provide information about the movement of the user. If the device 1 is not being worn, the accelerometer 9 will sense that there is little to no movement, and this be transmitted to the processor in the wireless communication module 7. The processor can then switch the device to a low energy sleep mode in order to conserve battery power.

Furthermore, the readings from the accelerometer 9 can be combined with other data sources to give additional insight into breathing and speech data. The readings will give an indication of how the movement of the user relates to their stammer. This allows the control algorithms and manual changes to decouple or differentiate between increased heart rate caused by exercise and increased heart rate caused by stress.

The processor and wireless transmitter/receiver 7 receives inputs from the microphones 4, the chest impedance and heart rate sensor 5, the strain and load gauge(s) 8, and the accelerometer 9. Software in the processor portion of the processor and wireless communication module 7 monitors all of the inputs, and is configured to provide an output if the sensed inputs fall outside certain pre-set parameters or ranges - that is, baseline speech and breathing patterns. The associated software can be integral with the transmitter/receiver 7 as above, or the processor and wireless communication module 7 can be in contact (via wireless communication) with a remote unit, such as a mobile phone, tablet or similar, loaded with appropriate software in the form of an app, or similar. The processor and wireless communication module 7 is also configured so that it can transmit and receive data irrespective of the performance of the user - that is, as required.

As noted above, the processor portion of the processor and wireless communication module 7 records the data received from the sensors. This data is then either processed by this processor directly, or transmitted to an external device and processed remotely. The sensors are hardwire-connected to the processor and wireless communication module 7. However, the processor and wireless communication module 7 connects wirelessiy either to a computer or a smart-phone, or a similar device.

As also noted above, the sensors can be contained wholly within the box 2, or distributed between the belt 3 and box 2, depending on their particular requirements. In some cases, a sensor may be held by both: for example the chest impedance and heart rate sensor 5 is configured in some embodiments so that the chip portion is located within the box 2, but the electrodes are located in the belt 3.

Use

A user wears the device in the same or a similar manner to that shown in figure 2. The sensors measure the data from the user as outlined above, and transmit this to the processor and wireless communication module 7.

In the preferred embodiment, full processing of the data is carried out remotely, via an app, or software located in another suitable, remotely-located device. A microcontroller or micro-computer in the wireless communication module 7 carries out an initial, basic, assessment of the data, and then transmits this in real time to a smartphone pre-loaded with an app. The app enables the display, processing and storage of all the measured data, such as chest expansion (via the strain gauge 8), heart rate (via the heart sensor 5), and words per minute (via the microphone 4).

The software is configured to process the data and provide real-time feedback to a user whenever the user is not breathing correctly.

Ideally, the device would measure the volume of air in the lungs, as a direct measurement. In practice, this is not possible, as it would require knowing the exact lung volume. Instead, two measures are used in combination:

• The tension that the chest applies to the belt is measured through the strain gauge/load cell 8. As the chest of the user expands, the belt suffers a strain and the load cell/strain gauge(s) 8 detect this.

• Chest impedance is calculated using the chest impedance and heart rate sensor 5 in the belt, connected to a circuit in the box to measure the impedance in the chest and to calculate an approximation of the chest expansion (chest expansion relates to impedance, so the greater the chest expansion, the higher the impedance). The same circuit and electrodes are used to measure the heart rate through a different function of the chip. A combination of both of these readings is used. However, they could be used separately, and it is considered that the chest impedance calculation from the chest impedance and heart rate sensor is more accurate. These readings could also be used in combination with other readings such as the acceleration/motion data from the accelerometer 9, or signals from the microphones, in order to improve the robustness of the data and/or calculation.

The inhalation/exhalation is monitored, and the software can assess when the breaming pattern falls outside certain pre-set parameters, or when the user is not practising habits which suppress stuttering, such as a low pitched voice, preferred words per minute, and speaking with enough air in the lungs. The software also provides feedback when there has not been stuttering "with control' for a long period of time (i.e. stammering in purpose have not been used).

The software causes the haptic actuator/transducer 6 to activate whenever the pattern is outside the pre-set parameters, or the pre-set parameters are exceeded, indicating that the user is displaying behaviour patterns liable to cause a stutter. The vibration caused by the haptic actuator/transducer 6 is felt by a user on their chest, and provides a warning to a user, allowing them to alter their behaviour as appropriate. In the preferred embodiment, different vibration patterns are used to send different warning signals - for example, if a user is speaking too fast, then three spaced-apart 'bursts' vibration are felt, followed by a short pause, and then another three bursts, the pattern continuing until a user has slowed their speech. If the inputs to the system indicate that a user has short breathing, and is not inhaling fully, then a series of regular taps' are created, with a frequency of around 40 per minute.

The intensity of the reminder can be set by the user via a control panel on the app, either to increase the user's awareness of the warnings, or minimise their intrusiveness, depending on preference.

While the app provides pre-set recommendations, targets can also be customised by the user in order make the learning experience more comfortable, or set more aggressive goals, again depending on preference. The user can set the limits on the app (for example: breaths per minute, chest expansion and breathing rate, pitch voice, word/syllable frequency or speech rate during speech) so that the app has their own preferences to assist them with correct monitoring and correction of their habits. The device is tuneable for user control. Data is stored in the app to allow performance history to be monitored. This provides a valuable source of motivation for the user, and allows them to tweak or alter the settings as appropriate.

The feedback is provided in real-time. This allows a user to be coached in real-time (i.e. during everyday activities), and thus assists a user with developing positive habits, and to reinforcing behaviours which reduce the probability of stuttering.

A device 1 as described provides feedback to a user in real-time. Real-time feedback is critical to acquire new (breathing, pausing-in-speech, low pitch voice and stuttering in purpose) habits. A device 1 such as the device described is also wearable during everyday activities. People who stutter do not stutter always, and therefore they tend to use those new techniques only in difficult speaking situations, outside of their comfort zone. However, these techniques are ineffective if only used when fear is high. A device such as device 1 can be comfortably worn at all times, and a user does not have to continuously put on and remove the device. They can wear the device all day, every day, including for low-pressure situations in which they are comfortable, and in anticipated and unanticipated high-pressure situations. This allows a user to develop good habits and to have the confidence that the device will provide support as required, even in unexpected situations.

A device as described above therefore promotes useful habits in order to assist a user at all times.

In addition the following features can be enabled as the device acquires data during use.

Associated software, such as that of the app, is configured to use initial baseline speech and breathing pattern settings to identify how a user breathes and speaks. As it acquires data during use, the software app will isolate sound recordal only (i.e. what is sensed using only the microphone), and identify common speech patterns over time from sound recordal only - that is, common elements such as increasing frequency of syllables and a greater number of words in between breaths for a particular user's particular speech pattern as the user starts to move towards a disordered speech pattern. Once the software has acquired sufficient data, it can identify patterns for a particular user based solely on sensed sound. This allows a user to receive a bespoke optimisation of the threshold settings and, eventually, to upgrade to a version of the device that only uses speech rather than speech and breathing patterns. Determining that the voice being processed belongs to the user, and not someone they are in conversation with or another person nearby, can be problematic. This can be addressed in one or more of the following ways:

• The microphone 4a (facing inwards towards the chest of a user) and

accelerometer 9, and the microcontroller/processor in the wireless

communication module 7, are used to record the voice of the user via their body. Software in the microcontroller/processor, or in the app, is used to compare this recording with the sound transmitted through the air in real-time, using either or both of the microphone in the mobile device on which the app is loaded, or the outwards-facing microphone 4b, so that the voice of the user can be isolated.

• The sounds picked up by the microphones 4a, 4b can be compared and

contrasted in real time by the microcontroller/processor in the wireless communication module 7, and the differences can be used to determine the direction of the source of sounds, and hence amplify and select solely the voice of the user.

• The user's voice can be recorded via a single microphone, and compared with the known voice of the user/the user's voice spectrum (this will have been previously attained and implemented through the training of a machine learning algorithm). A filter is applied, so that their voice can be selected from background noise.

• Multiple microphones (e.g. microphones 4a and 4b) can be used, with the microcontroller/processor in the wireless communication module 7 performing an independent component analysis' to determine and extract individual sources of noise, and in doing so determining the voice of the user.

The associated software can be located remotely (e.g. in the app loaded onto the mobile device), with the device 1 communicating in real-time or updating frequently, or at intervals (such as once every twenty-four hours - e.g. at the end of the day).

The device 1 can also collect information on the user's physiological state with time. This information can be sent anonymously to a pool data centre. This will allow broader statistical research on stuttering, to increase understanding of the reasons for which people stutter, and improve diagnosis and treatment methods, and also improve the effectiveness of the device. The system can also be set to provide an alert or real time feedback whenever the pitch of a user's voice starts to increase. That is, when a user is not using a 'deep and breathy tone' (a low pitch voice). A baseline voice pitch for a user is set, and if the frequency/pitch starts to move outside a set range from the established baseline, then an alert is provided. This is useful as speaking with a low-pitched voice has been shown to be useful for people who stutter.

The aim of the electronic device is to increase the fluency of people by encouraging them to use the full capacity of the diaphragm when breathing, limiting the words spoken per breath and per minute, using a low pitch of voice and stammering on purpose by following a breathing and speaking technique where the person is in control of their disfluency. A breathing pattern of less than seven breathes per minute is beneficial in aiding relaxation, and in the case of people who stutter assists with their fluency.

A particular embodiment of device has been described above. In variations, the device can use electrodes and an impedance sensor (to monitor chest impedance), or could use only the microphone (to monitor the sound of the lungs) to monitor the breathing of a user, instead of a strain gauge. This will allow the size of the device to potentially be reduced, and may allow the device to be secured directly to the user's chest via adhesion to the skin, removing the need for a chest belt to be used. As for the first embodiment, the software provides real-time feedback to a user whenever the user is not breathing correctly (e.g. the chest impedance sensed by the electrodes impedance sensor, and/or the breathing pattern sensed by the

microphone or microphones). This data is monitored by the associated software in a similar manner to that described above for the first embodiment. The microphone, in combination with the software, can also be used to quantify the breathing patterns (i.e. number of breaths per minute, inhalation and exhalation time), in addition or instead of what is described for the first embodiment above. The breathing and lung, and number of breaths per minute can be monitored using the microphone and software, and this can be compared to data sets for 'correct¹ lung sound (for a given weight age and gender. If necessary, a user can be alerted, and real-time feedback can be provided, whenever the user is not breathing correctly.

It should also be noted that in the embodiment above, the microphone 4, the heart rate sensor 5, the strain and load gauge 8, the processor and wireless

transmitter/receiver 7 and the accelerometer 9 are all located in the box 2, which in use is located in the centre of the user's chest These elements could instead be distributed around the belt 3 rather than in one single location.

It should also be noted that the inductance of the chest cavity can be an important measurement as it gives an indication of the depth and rate of the user's breath. The chest cavity readings are data that is recorded, either through the deflection of the strain and load gauge or gauges 8 as the chest cavity expands, or via a

measurement of the changing inductance. In practice, for robustness, both are likely to be used together. Measuring chest inductance can be less invasive, and as inductance can be measured using sensors that do not require moving/deforming parts, this may be preferable in certain circumstances and may offer a better long- term solution.

All of the methods described above may involve a comparison of particular metrics and therefore a comparison to pre-set or pre-recorded "baseline values".

Alternatively the methods may involve a more complex comparison of patterns, for example via a neural network.

As described in the example above, the device and methodology are for assistance with increasing the fluency of people who stutter. The device and methodology could also be used for assisting people who suffer from dysphonia - that is, difficulty in speaking due to a physical disorder of the mouth, tongue, throat, or vocal cords. Some people with dysphonia are only able to articulate a few words before becoming hoarse. Others, such as singers, teachers or professional speakers can also frequently suffer from dysphonia. The device and methodology are also of use in alleviating this condition. The treatment for dysphonia is very similar to that for stammering, with a focus on 'speech-related breathing' practices.

The device and methodology could also be used to assist with the treatment of hyperventilation, dysphonia or anxiety management.

Claims

1. An electronic fluency device, comprising:

at least one sensor configured to measure biometric data relating to the speech of a user

a processor configured to receive input from the at least one sensor, and to assess the input against baseline speech and breathing patterns;

a warning mechanism configured to provide an alert to a user, the warning mechanism configured to receive a controi signal from the processor to initiate the alert;

the processor further configured to transmit the control signal if the biometric data falls outside baseline values.

2. An electronic fluency device as claimed in claim 1 wherein in use, the at least one sensor is configured to be worn by a user.

3. An electronic fluency device as claimed in claim 2 further comprising a chest belt configured to extend around the chest of a user.

4. An electronic fluency device as claimed in claim 3 further comprising a box, the chest belt connected to and extending from the box, at least one of the sensor or sensors located in the box.

5. An electronic fluency device as claimed in claim 3 or claim 4 wherein at least one of the sensor or sensors is configured to locate on the belt.

6. An electronic fluency device as claimed in any one of claims 2 to 5 wherein the sensor or sensors is/are configured to measure costal rib expansion and contraction.

7. An electronic fluency device as claimed in claim 6 wherein the sensor or sensors comprises a strain gauge/load cell.

8. An electronic fluency device as claimed in any one of claims 1 to 7 wherein the sensor or sensors comprises a chest impedance and heart rate sensor configured to assess chest impedance.

9. An electronic fluency device as claimed in any one of claims 1 to 8 wherein the sensor or sensors comprises at least one microphone.

10. An electronic fluency device as claimed in claim 9 wherein the at least one microphone comprises at least one inwards-facing microphone.

11. An electronic fluency device as claimed in claim 9 or claim 10 wherein the at least one microphone comprises at least one outwards-facing microphone.

12. An electronic fluency device as claimed in any one of claims 1 to 11 wherein the sensor or sensors comprises an accelerometer.

13. An electronic fluency device as claimed in any one of claims 1 to 12 wherein the warning mechanism comprises a haptic actuator transducer configured to provide a vibration against the chest of a user when activated.

14. An electronic fluency device as claimed in any one of claims 1 to 13 wherein the processor comprises a programmable device.

15. An electronic fluency device as claimed in claim 14 wherein the processor comprises a microcontroller.

16. An electronic fluency device as claimed in claim 14 or claim 15 wherein the programmable device is integral with the sensor.

17. An electronic fluency device as claimed in claim 14 or claim 15 wherein the programmable device comprises a remotely located programmable device, in communication with the sensor via wireless communication.

18. An electronic fluency device as claimed in claim 17 wherein the remotely-located programmable device is a mobile device.

19. An electronic fluency device as claimed in any one of claims 16 to 18 wherein the programmable device is configured to receive readings from the sensor or sensors and compare these to preset baseline readings.

20. An electronic fluency device as claimed in claim 19 wherein the programmable device is configured so that the preset baseline readings can be adjusted.

21. An electronic fluency device as claimed in any one of claims 14 to 20 wherein the processor is further configured to store and/or transmit data relating to one or both of breathing and speech patterns.

22. An electronic fluency device as claimed in any one of claims 1 to 21 wherein the control signal comprises multiple signals, each signal causing the warning mechanism to provide a different alert, the different alerts indicative of a particular pattern of speech or breathing falling outside baseline speech and breathing patterns.

23. A method of monitoring biometric data and speech patterns for assisting with speech therapy, comprising the steps of:

i) measuring biometric data relating to the speech of a user;

24. A method of monitoring biometric data and speech patterns for assisting with speech therapy wherein in the step of measuring biometric data, the biometric data consists of one or more of: costal rib expansion and contraction; chest impedance; the voice of the user; ambient external background noise; heart rate.

25. A method of monitoring biometric data and speech patterns for assisting with speech therapy wherein in the step of comparing measured biometric data against baseline speech and breathing patterns, the user's voice is monitored, and patterns for the user are identified, the alert provided if the patterns move towards a disordered speech pattern.

26. A method of monitoring biometric data and speech patterns for assisting with speech therapy wherein the user's voice is specifically monitored for an increasing number of words between breaths, and/or an increasing frequency of words for a set time period.

27. A method of monitoring biometric data and speech patterns for assisting with speech therapy as claimed in claim 25 or claim 26 wherein the user's voice is monitored using a user-specific microphone, and ambient noise is monitored using a separate microphone, the difference between the two used to isolate the

characteristics of the user's voice.