Field

The present invention relates to an apparatus for assistive communication.

Background

Many modern economies are facing the issue of rapidly ageing populations, and there are no easy ways to tackle this problem. In connection, research have found that most adults would prefer to age in place, a concept known as "age in place" - that is, to remain in their home of choice as long as possible. For example, a study in America reports that over 90% of adults aged 65 and above prefer to stay in their current residences as they age, which is hardly surprising as one may empathized. However, for older adults, it can be a risk for them to be at home alone (when their children are away), since there exist many injury threats to the older adults in the common household. In event that the older adults suffer an injury (e.g. had a fall), or are in need of some immediate assistance, when the children are not around, it can present a dangerous life situation for these older adults without someone around to render timely assistance. Needless to say, this is highly undesirable, and will also impact upon the older adults' capability to successfully age in place.

One object of the present invention is therefore to address at least one of the problems of the prior art and/or to provide a choice that is useful in the art.

Summary

According to a 1 ^st aspect, there is provided an apparatus for assistive communication, comprising: a base device configured to process voice commands by matching with at least one predetermined phrase stored in a database, the database being arranged locally in the base device; and at least one sensor device configured to receive the voice commands from a user, and to transmit the received voice commands to the base device for processing, wherein if the voice commands match the at least one predetermined phrase in the database, the base device is activated to communicate with an external third party for response.

Beneficially, for processing the voice commands, an internet connection is not required, since the voice commands are processed directly by the base device, with reference to the locally installed database. Therefore, the processing can be performed much faster and timely.

Preferably, the at least one sensor device may include a plurality of sensor devices configured to be spatially positioned for receiving the voice commands.

Preferably, the apparatus may further include an activation device configured to be actuated to directly activate the base device to communicate with the external third party, without need to provide the voice commands.

Preferably, the activation device may be a wearable device.

Preferably, the base device may further comprise a sensor device integrated therewith for receiving the voice commands directly by the base device from the user.

Preferably, the base device may be configured to communicate with the external third party using an internet connection. Preferably, the sensor device may include an audio sensor.

Preferably, the sensor device may be configured to encrypt the received voice commands, prior to transmitting the voice commands to the base device, and the base device is therefore configured to decrypt the received encrypted voice commands, prior to processing the voice commands.

According to a 2 ^nd aspect, there is provided a method performed by an apparatus for assistive communication, where the apparatus includes at least one sensor device, and a base device locally arranged with a database. The method comprises: receiving voice commands by the sensor device from a user; transmitting the received voice commands by the sensor device to the base device for processing; processing the voice commands by the base device by matching with at least one predetermined phrase stored in the database; and activating the base device to communicate with an external third party for response, if the voice commands match the at least one predetermined phrase in the database.

It should be apparent that features relating to one aspect of the invention may also be applicable to the other aspects of the invention.

These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter.

Brief Description of the Drawings

Embodiments of the invention are disclosed hereinafter with reference to the accompanying drawings, in which:

FIG. 1 shows schematics of an apparatus for assistive communication, according to an embodiment;

FIG. 2 is a block diagram of a sensor device of the apparatus of FIG. 1 ;

FIG. 3 is a block diagram of a base device of the apparatus of FIG. 1 ; and

FIG. 4 is a flow diagram of a corresponding method performed by the apparatus of FIG. 1.

Detailed Description of Preferred Embodiments

An apparatus 100 for assistive communication is herein disclosed, according to an embodiment, as shown in FIG. 1. Broadly, the apparatus 100 comprises a base device 102 (i.e. also known as a processor unit) configured to process voice commands by matching with at least one predetermined phrase stored in a database (not shown), the database being arranged locally in the base device 102; and at least one sensor device 104 (i.e. also known as a sensor unit) configured to receive the voice commands from a user, and to transmit the received voice commands to the base device 102 for processing. If the base device 102 determines that the voice commands match the at least one predetermined phrase in the database, the base device 102 is then activated to communicate with an external third party (not shown) for response. It is to be appreciated that the definition of the at least one sensor device 104 includes a plurality of such sensor devices 104 configured to be spatially positioned for receiving the voice commands, which beneficially provides multiple receiving channels to improve accuracy of pickup of the voice commands issued by the user. Under this configuration, voice streams from the different sensor devices 104 transmitted to the base device 102 are processed sequentially by the base device 102. Specifically, the base device 102 processes a chunk of voice stream from a first sensor device 104 (in the different sensor devices 104), followed by those from a second sensor device 104, so on and so forth, until all the voice streams have been processed. When a match (of the voice commands with at least one predetermined phrase in the database) is found, subsequent matches within a predetermined time period are then ignored. In one example, in FIG. 1 , four of the sensor devices 104 are utilised, but it is not to be construed as limiting in any way, since as many of the sensor devices 104 as needed can be deployed, depending on an application scenario intended (e.g. the size of a room/space in which the apparatus 100 is to be setup). Each sensor device 104 has a predefined audio sensor range for detecting and capturing the voice commands. Also, the base device 102 and sensor devices 104 are conveniently arranged to be "plug-and-play", because deployment is simply effected by plugging the base device 102 and sensor devices 104 respectively into wall power outlets (or suitable power sources), and thereafter switching on the outlet power to start operating the base device 102 and sensor devices 104.

Specifically, the apparatus 100 is devised to help users (e.g. elderly adults) in needy situations where no communication devices are within physical reach to enable those users to reach out/call for assistance. The apparatus 100 is configured to particularly "listen" for voice commands from the users, via the sensor devices 104, and to respond to pre-programmed key phrases, as such as "ALERT! HELP!". Upon successfully detecting the correct key phrases in the voice commands (by matching with one or more predetermined phrases stored in the database), the base device 102 transmits a response (e.g. via text messages or other forms of suitable notifications) to the external third party to call for assistance or related services. In certain embodiments, it is envisaged that users of the apparatus 100 may also be able to hold a two-way communication with external third parties using the base device 102, and the sensor devices 104; the apparatus 100 is configurable to enable bidirectional communication between the users and external third parties.

FIG. 2 is a block diagram of the sensor device 104, which includes an audio sensor 202 (e.g. a microphone), a parabolic dish 204, an analogue-to-digital signal converter (ADC) 206, an encryptor 208, a transceiver 210, an AC-to-DC power converter 212, and other miscellaneous features (e.g. a speaker, operating buttons, light arrays and etc.). Specifically, sound waves (of voice commands issued by a user) which impinge on the sensor device 104, are first concentrated/focused by the parabolic dish 204 for re-direction to the audio sensor 202. The sound waves received by the audio sensor 202 are then converted into corresponding electric signals, and thereafter passed to the ADC 206 for further processing into associated digital signals. The digital signals generated are (optionally) encrypted by the encryptor 208 prior to being transmitted as RF signals over the air (e.g. using Bluetooth/Wi-Fi) by the transceiver 210 to the base device 102. As will be appreciated, the power converter 212 converts AC line voltage from a wall power socket into DC power for operating various components of the sensor device 104 as set out above. FIG. 3 is a block diagram of the base device 102, which includes a transceiver 302, a decryptor 304, a processor 306, a modem/Wifi transmitter 308, a SIM card slot (not shown), an AC-to-DC power converter 310, and other miscellaneous features (e.g. a speaker, operating buttons, light arrays, an optional audio sensor, and antennas for Bluetooth/Wi-Fi transmission). Also, a storage device (not shown) is also present on the base device 102 to store the database (e.g. a computer file(s)) of the predetermined phrases, as well as to store a speech recognition engine (implemented as a computer software), which is required when processing the received voice commands from the sensor device 104. RF signals received by the transceiver 302 (transmitted from the sensor device(s) 104) are first converted into baseband digital signals and provided to the decryptor 304, provided the converted digital signals are in encrypted form; otherwise the converted digital signals need not be provided to the decryptor 304. The decryptor 304 then generates the original audio signal (corresponding to the received voice commands) by decrypting the encrypted digital signals, and the original audio signals obtained are provided to the processor 306 for processing. In particular, the processor 306 processes the audio signals (by matching with predetermined phrases stored in the database) to determine presence of any pre-programmed key phrases that may be contained within the received voice commands. Upon successfully detecting presence of pre-programmed key phrase(s) within the received voice commands, the base device 102 is arranged to send a notification to external third parties for assistance via the modem/Wifi transmitter 308 or an internet connection (which may be realised using a mobile broadband connection via the SIM card slot). Separately, the power converter 310 converts AC line voltage from a wall power socket into DC power for operating various components of the base device 102.

Next, an associated method 400 performed by the disclosed apparatus 100 for assistive communication is shown in FIG. 4. The method 400 comprises: at step 402, receiving voice commands by the sensor device 104 from a user; at step 404, transmitting the received voice commands by the sensor device 104 to the base device 102 for processing; at step 406, processing the voice commands by the base device 102 by matching with at least one predetermined phrase stored in the database; and at step 408, activating the base device 102 to communicate with an external third party for response, if the voice commands match the at least one predetermined phrase in the database.

To further clarify step 406, when the voice commands (i.e. as digitized sound signals) are received by the base device 102 (from the sensor device 104), the base device 102 first converts the digitized sound signals into corresponding commands (in text format) using the speech recognition engine, and an associated suitable database is utilised by the speech recognition engine to assist with the conversion. Next, the base device 102 compares the converted commands using a customised computer program (configured to perform the matching/comparison) with at least one predetermined phrase in the database that stores the plurality of predetermined phrases (i.e. the database may be a computer file that stores all the predetermined phrases recognizable by the base device 102). It is also to be appreciated that the predetermined phrase may alternatively be pre-stored in the customised computer program, and so the database may not be needed in such cases. If the predetermined phrase matches any of the converted commands, then the base device 102 is consequently activated to communicate with the external third party at step 408.

To summarise, the proposed apparatus 100 is able to process the received voice commands locally on the base device 102, without need to depend on online backend servers for the processing, because the database itself is stored locally on the base device 102 (either within the customised computer program or the computer file). That means in this respect, processing of the received voice commands is performed without an internet connection, and hence advantageously is much faster and timely compared to conventional solutions. Further, by using multiple sensor devices 104 (if necessary), pickup of the voice commands can be more accurate, since multiple receiving channels are spatially available simultaneously to assist with detection of voice commands issued. While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary, and not restrictive; the invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practising the claimed invention.

For example, the apparatus 100 also may further include an activation device (e.g. implemented as a wearable device such as a watch, or alternatively as a remote control) which is configured to be actuated (e.g. by pressing a button on the activation device) to directly activate the base device 102 to communicate with the external third party, without need to provide the voice commands. It is to be appreciated that actuation of the activation device is not restricted to merely actuation by hand alone. Indeed, having the activation device is useful in instances such as where the sensor device(s) 104 is not able to correctly and accurately pick up the voice commands for transmission to the base device 102.

Additionally, the base device 102 may further incorporate a sensor device 104 within in order to receive voice commands directly by the base device 102 from the user, and this beneficially adds another receiving channel for picking up voice commands. In one example, the base device 102, incorporated with the sensor device 104, can be utilised independently by itself without deploying separate sensor devices 104, as afore described with reference to FIG. 1 .

Yet alternatively, the database in the base device 102 is customisable to add new phrases to the database, or delete/edit existing phrases stored in the database. Also, usage of the encryptor 208 in the sensor device 104, and the decryptor 304 in the base device 102 may be optional in certain variant embodiments.