where M is the mass of perfume (material) per pulse volume, D represents the diffusivity of the perfume, and f() is a function representing diffusion of the perfume (material) in 3D space.  In embodiments, the perfume (material) may be substantially undiluted in the perfume reservoir. 

 

Calculations have established that, surprisingly, when the radius is less than 100cm the dispensed volume may be extremely small to maintain the scent (perfume) bubble, in embodiments less than one nanolitre.  Thus, very surprisingly, a wearable device of the type described above may, with a modest scent reservoir of around 1 millilitre, potentially create and maintain a scent bubble for a period of a year or longer.  The target radius may be determined dependent upon the location of the wearable device and may be smaller when the wearable is located on the head rather than on the chest or wrist.

 

As described further later, the method may include detecting one or more speech characteristics of the wearer and creating and maintaining the personalised bubble of scent in response.

 

The wearable device may have a plurality of perfume reservoirs. 

 

A user interface may be provided to enable a user to control the scent.  In response to user scent control data from the user interface the method may select one or more perfume reservoirs from the plurality of perfume reservoirs and/or controlling a pulse volume and/or pulse interval for the selected reservoir(s), in order to control a scent of the scent bubble dependent upon user control from the user interface.

 

Different perfume reservoirs may hold different perfume ingredients.  Then different pulse volumes and/or pulse intervals may be used to control dispensing from the different perfume reservoirs in order to compensate for different respective dispersal rates of the perfume ingredients in the reservoirs.

 

Whether or not a scent bubble is created and maintained a method may include inputting context data relating to a use context of the wearable device and providing the context data to a dispensing/scent control neural network.  The dispensing/scent control neural network outputs dispensing/scent control data in response to the context data.  A liquid/scent dispensed by a wearable device is controlled dependent upon the dispensing/scent control data.  The context data may comprise, for example: time data, day data, location data, activity data, voice data, music data e.g. music genre/sentiment data, scheduled or other contemporaneous event data; or other data.

 

The neural network may have previously been trained based on data collected from a user’s operation of the device and/or data from multiple users may be pooled for training the neural network.  The neural network may alternatively be provided pre-trained for use in a wearable device.  Such a neural network may be a copy of a neural network that has been trained elsewhere.

 

In a related aspect the invention provides a wearable device comprising: a liquid reservoir; a liquid dispenser coupled to the liquid reservoir; and a controller to dispense a volume of liquid from the device at intervals; wherein said controller is configured to control said wearable device to dispense a pulse volume of liquid from said liquid reservoir at a pulse interval. 

 

In some implementations the pulse volume and the pulse interval are controlled to maintain a mass per unit volume concentration of said liquid above a threshold level within a defined radius from said wearable device. 

 

The liquid may comprise perfume.  Other liquids which may be dispensed include, but are not limited to:  essential oils; decongestants; deodorant or antiperspirants (e.g. for hyperhidrosis); insect repellent; hormones; pharmaceuticals; human pheromones; CBD oil / cannabidiol; cognitive-enhancing drugs, nootropics and "micro-dosing" psychedelic medicines (e.g. LSD, psilocybin) for therapeutic applications; food or other flavour replicating liquids; personal care active ingredients; moisturising liquid; sun protection liquid; anti-cellulite liquid; vitamin-enhanced liquid; anti-tobacco liquid; anti-UV/anti-pollution liquid; and hydrating mist, e.g. water.

 

In some implementations the wearable device has an acoustic transducer - for example, a microphone may be built into the device or the device may be wired or wirelessly coupled to another device with an acoustic transducer such as a mobile phone.  The acoustic transducer may be coupled to a processor to process a speech signal of a wearer of the wearable device to detect one or more speech characteristics of the wearer.  The signal processing may be local, for example, performed in the wearable device or on an associated device such as the mobile phone, or remote, for example, performed on a remote server. 

 

A result of the processing, more particularly detection of the one or more speech characteristics may be that a signal is provided for the controller to dispense one or more pulses of liquid.  The liquid may be a perfume as previously described or some other liquid.  For example, the system may be configured to detect stress in a voice; the system may then dispense stress-reducing perfume such as lavender. 

 

In some implementations, one or more speech characteristics may comprise one or more speech characteristics of mania or depression.  The processor may thus be configured to process the speech signal to classify the wearer as in a manic or depressive state.  A result of the processing, more particularly detection of the mania or depression, may again be that a signal is provided for the controller to dispense one or more pulses of liquid.  The liquid may be a stress-reducing perfume such as lavender, or in other implementations, the liquid may comprise a drug for treatment of the mania and/or depression.

 

Thus in a related innovative aspect, there is provided a method of calming a person.  The method may comprise providing the person with a wearable device comprising a liquid reservoir, a liquid dispenser coupled to the liquid reservoir, and a controller to dispense a volume of liquid from the wearable device at intervals.  The method may further comprise capturing speech data from the person as the person is speaking.  The method may further comprise processing the captured speech data to detect one or more characteristics of the speech and, in response, to the detected characteristic.  The method may further comprise controlling said wearable device to dispense a pulse volume of liquid from said liquid reservoir at pulse intervals.  The method may further comprise controlling said pulse volume and said pulse interval to maintain a mass per unit volume concentration of said liquid above a threshold level within a defined radius from said wearable device. 

 

Processing the captured speech data may comprise processing the speech data to detect a state of mania or depression in the person from a voice of the person, for example from a speed and/or one of speech of the person. The method may then further comprise creating and maintaining the personalised bubble of liquid in response to the detected state of mania or depression.

 

In a further aspect there is provided a wearable device comprising a plurality of perfume reservoirs each for a respective perfume or perfume ingredient; at least one dispenser coupled to the perfume reservoirs; and a controller to dispense perfume or a perfume ingredient from the wearable device at intervals.  The controller may be configured to control the wearable device to dispense a pulse volume of the perfume or perfume ingredient from each perfume reservoir at a respective pulse interval.

 

In some implementations, the device has a user interface to enable a user to control the scent.  In response to user scent control data from the user interface the device may select one or more reservoirs from the plurality of reservoirs and/or control a pulse volume and/or pulse interval for the selected reservoir(s).

 

In some implementations, where the reservoirs are for perfume ingredients, the device is configured to use different pulse volumes and/or pulse intervals to control dispensing from the reservoirs in order to compensate for different respective dispersal rates of perfume ingredients in the reservoirs

 

In some implementations, the wearable device includes, or is coupled to, a dispensing/scent control neural network (or other machine learning system).  The dispensing/scent control neural network (or other machine learning system) may be configured to receive context data relating to a use context of the wearable device, and in response to output dispensing/scent control data for controlling a liquid/scent dispensed by a wearable device.

 

A wearable device with multiple reservoirs as described above may be configured/controlled to create, and preferably maintain, a scent bubble as previously described.

 

Features of the previously described aspects and embodiments of the invention may be combined with the above methods.

 

The invention further provides processor control code to implement the above-described methods, for example on a general purpose computing device, or on a mobile device, or on a digital signal processor (DSP).  The code is provided on a non-transitory physical data carrier such as a disk, CD- or DVD-ROM, programmed memory such as non-volatile memory e.g. Flash, or read-only memory (Firmware).  Code and/or data to implement embodiments of the invention may comprise source, object or executable code in a conventional programming language (interpreted or compiled) such as C, or assembly code, or code for a hardware description language.  As the skilled person will appreciate such code and/or data may be distributed between a plurality of coupled components in communication with one another. 

​

Examples of liquids: