U.S. Pat. No. 11,554,322

DETAILED DESCRIPTION

This disclosure relates generally to computer ecosystems including aspects of consumer electronics (CE) device networks such as but not limited to distributed computer game networks, augmented reality (AR) networks, virtual reality (VR) networks, video broadcasting, content delivery networks, virtual machines, and artificial neural networks and machine learning applications.

A system herein may include server and client components, connected over a network such that data may be exchanged between the client and server components. The client components may include one or more computing devices including AR headsets, VR headsets, game consoles such as Sony PlayStation® and related motherboards, game controllers, portable televisions (e.g. smart TVs, Internet-enabled TVs), portable computers such as laptops and tablet computers, and other mobile devices including smart phones and additional examples discussed below. These client devices may operate with a variety of operating environments. For example, some of the client computers may employ, as examples, Orbis or Linux operating systems, operating systems from Microsoft, or a Unix operating system, or operating systems produced by Apple, Inc. or Google. These operating environments may be used to execute one or more programs/applications, such as a browser made by Microsoft or Google or Mozilla or other browser program that can access websites hosted by the Internet servers discussed below. Also, an operating environment according to present principles may be used to execute one or more computer game programs/applications and other programs/applications that undertake present principles.

Servers and/or gateways may include one or more processors executing instructions that configure the servers to receive and transmit data over a network such as the Internet. Additionally, or alternatively, a client and server can be connected over a local intranet or a virtual private network. A server or controller may be instantiated by a game console and/or one or more motherboards thereof such as a Sony PlayStation®, a personal computer, etc.

Information may be exchanged over a network between the clients and servers. To this end and for security, servers and/or clients can include firewalls, load balancers, temporary storages, and proxies, and other network infrastructure for reliability and security. One or more servers may form an apparatus that implement methods of providing a secure community such as an online social website or video game website to network users to communicate crowdsourced in accordance with present principles.

As used herein, instructions refer to computer-implemented steps for processing information in the system. Instructions can be implemented in software, firmware or hardware and include any type of programmed step undertaken by components of the system.

A processor may be any conventional general-purpose single- or multi-chip processor that can execute logic by means of various lines such as address lines, data lines, and control lines and registers and shift registers.

Software modules described by way of the flow charts and user interfaces herein can include various sub-routines, procedures, etc. Without limiting the disclosure, logic stated to be executed by a particular module can be redistributed to other software modules and/or combined together in a single module and/or made available in a shareable library.

As indicated above, present principles described herein can be implemented as hardware, software, firmware, or combinations thereof; hence, illustrative components, blocks, modules, circuits, and steps are set forth in terms of their functionality.

Further to what has been alluded to above, logical blocks, modules, and circuits described below can be implemented or performed with a general-purpose processor, a digital signal processor (DSP), a field programmable gate array (FPGA) or other programmable logic device such as an application specific integrated circuit (ASIC), discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A processor can be implemented by a controller or state machine or a combination of computing devices.

The functions and methods described below may be implemented in hardware circuitry or software circuitry. When implemented in software, the functions and methods can be written in an appropriate language such as but not limited to Java, C# or C++, and can be stored on or transmitted through a computer-readable storage medium such as a random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), compact disk read-only memory (CD-ROM) or other optical disk storage such as digital versatile disc (DVD), magnetic disk storage or other magnetic storage devices including removable thumb drives, etc. A connection may establish a computer-readable medium. Such connections can include, as examples, hard-wired cables including fiber optics and coaxial wires and digital subscriber line (DSL) and twisted pair wires. Such connections may include wireless communication connections including infrared and radio.

Components included in one embodiment can be used in other embodiments in any appropriate combination. For example, any of the various components described herein and/or depicted in the Figures may be combined, interchanged or excluded from other embodiments.

“A system having at least one of A, B, and C” (likewise “a system having at least one of A, B, or C” and “a system having at least one of A, B, C”) includes systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and/or A, B, and C together, etc.

Now specifically referring toFIG.1, an example system10is shown, which may include one or more of the example devices mentioned above and described further below in accordance with present principles. The first of the example devices included in the system10is a consumer electronics (CE) device such as an audio video device (AVD)12such as but not limited to an Internet-enabled TV with a TV tuner (equivalently, set top box controlling a TV). However, the AVD12alternatively may be an appliance or household item, e.g. computerized Internet enabled refrigerator, washer, or dryer. The AVD12alternatively may also be a computerized Internet enabled (“smart”) telephone, a tablet computer, a notebook computer, an augmented reality (AR) headset, a virtual reality (VR) headset, Internet-enabled or “smart” glasses, another type of wearable computerized device such as a computerized Internet-enabled watch, a computerized Internet-enabled bracelet, a computerized Internet-enabled music player, computerized Internet-enabled head phones, a computerized Internet-enabled implantable device such as an implantable skin device, other computerized Internet-enabled devices, etc. Regardless, it is to be understood that the AVD12is configured to undertake present principles (e.g., communicate with other consumer electronics (CE) devices to undertake present principles, execute the logic described herein, and perform any other functions and/or operations described herein).

Accordingly, to undertake such principles the AVD12can be established by some or all of the components shown inFIG.1. For example, the AVD12can include one or more displays14that may be implemented by a high definition or ultra-high definition “4K” or higher flat screen and that may be touch-enabled for receiving user input signals via touches on the display. The AVD12may include one or more speakers16for outputting audio in accordance with present principles, and at least one additional input device18such as an audio receiver/microphone for entering audible commands to the AVD12to control the AVD12. The example AVD12may also include one or more network interfaces20for communication over at least one network22such as the Internet, an WAN, an LAN, etc. under control of one or more processors. Thus, the interface20may be, without limitation, a Wi-Fi transceiver, which is an example of a wireless computer network interface, such as but not limited to a mesh network transceiver. Furthermore, note the network interface20may be, e.g., a wired or wireless modem or router, or other appropriate interface such as, for example, a wireless telephony transceiver, or Wi-Fi transceiver as mentioned above, etc.

It is to be understood that the one or more processors control the AVD12to undertake present principles, including the other elements of the AVD12described herein such as controlling the display14to present images thereon and receiving input therefrom. The one or more processors may include a central processing unit (CPU)24as well as a graphics processing unit (GPU)25on a graphics card25A.

In addition to the foregoing, the AVD12may also include one or more input ports26such as, e.g., a high definition multimedia interface (HDMI) port or a USB port to physically connect (e.g., using a wired connection) to another consumer electronics (CE) device and/or a headphone port to connect headphones to the AVD12for presentation of audio from the AVD12to a user through the headphones. For example, the input port26may be connected via wire or wirelessly to a cable or satellite source26aof audio video content. Thus, the source26amay be, e.g., a separate or integrated set top box, or a satellite receiver. Or, the source26amay be a game console or disk player containing content that might be regarded by a user as a favorite for channel assignation purposes. The source26awhen implemented as a game console may include some or all of the components described below in relation to the CE device44and may implement some or all of the logic described herein.

The AVD12may further include one or more computer memories28such as disk-based or solid-state storage that are not transitory signals, in some cases embodied in the chassis of the AVD as standalone devices or as a personal video recording device (PVR) or video disk player either internal or external to the chassis of the AVD for playing back AV programs or as removable memory media. Also in some embodiments, the AVD12can include a position or location receiver such as but not limited to a cellphone receiver, GPS receiver and/or altimeter30that is configured to, e.g., receive geographic position information from at least one satellite or cellphone tower and provide the information to the processor24and/or determine an altitude at which the AVD12is disposed in conjunction with the processor24. However, it is to be understood that another suitable position receiver other than a cellphone receiver, GPS receiver and/or altimeter may be used in accordance with present principles to, for example, determine the location of the AVD12in all three dimensions.

Continuing the description of the AVD12, in some embodiments the AVD12may include one or more cameras32that may be, e.g., a thermal imaging camera, a digital camera such as a webcam, an infrared (IR) camera, and/or a camera integrated into the AVD12and controllable by the processor24to generate pictures/images and/or video in accordance with present principles. Also included on the AVD12may be a Bluetooth transceiver34and other Near Field Communication (NFC) element36for communication with other devices using Bluetooth and/or NFC technology, respectively. An example NFC element can be a radio frequency identification (RFID) element.

Further still, the AVD12may include one or more auxiliary sensors37(e.g., a motion sensor such as an accelerometer, gyroscope, cyclometer, or a magnetic sensor, an infrared (IR) sensor, an optical sensor, a speed and/or cadence sensor, a gesture sensor (e.g., for sensing gesture command), etc.) providing input to the processor24. The AVD12may include an over-the-air TV broadcast port38for receiving OTA TV broadcasts providing input to the processor24. In addition to the foregoing, it is noted that the AVD12may also include an infrared (IR) transmitter and/or IR receiver and/or IR transceiver42such as an IR data association (IRDA) device. A battery (not shown) may be provided for powering the AVD12.

Still referring toFIG.1, in addition to the AVD12, the system10may include one or more other consumer electronics (CE) device types. In one example, a first CE device44may be used to send computer game audio and video to the AVD12via commands sent directly to the AVD12and/or through the below-described server while a second CE device46may include similar components as the first CE device44. In the example shown, the second CE device46may be configured as an AR or VR headset worn by a user47as shown. In the example shown, only two CE devices44,46are shown, it being understood that fewer or greater devices may also be used in accordance with present principles.

In the example shown, all three devices12,44,46are assumed to be members of a network such as a secured or encrypted network, an entertainment network or Wi-Fi in, e.g., a home, or at least to be present in proximity to each other in a certain location and able to communicate with each other and with a server as described herein. However, present principles are not limited to a particular location or network unless explicitly claimed otherwise.

The example non-limiting first CE device44may be established by any one of the above-mentioned devices, for example, a smart phone, a digital assistant, a portable wireless laptop computer or notebook computer or game controller (also referred to as “console”), and accordingly may have one or more of the components described below. The second CE device46without limitation may be established by an AR headset, a VR headset, “smart” Internet-enabled glasses, or even a video disk player such as a Blu-ray player, a game console, and the like. Still further, in some embodiments the first CE device44may be a remote control (RC) for, e.g., issuing AV play and pause commands to the AVD12, or it may be a more sophisticated device such as a tablet computer, a game controller communicating via wired or wireless link with a game console implemented by another one of the devices shown inFIG.1and controlling video game presentation on the AVD12, a personal computer, a wireless telephone, etc.

Accordingly, the first CE device44may include one or more displays50that may be touch-enabled for receiving user input signals via touches on the display50. Additionally, or alternatively, the display(s)50may be an at least partially transparent display such as an AR headset display or a “smart” glasses display or “heads up” display, as well as a VR headset display, or other display configured for presenting AR and/or VR images.

The first CE device44may also include one or more speakers52for outputting audio in accordance with present principles, and at least one additional input device54such as, for example, an audio receiver/microphone for entering audible commands to the first CE device44to control the device44. The example first CE device44may further include one or more network interfaces56for communication over the network22under control of one or more CE device processors58. Thus, the interface56may be, without limitation, a Wi-Fi transceiver, which is an example of a wireless computer network interface, including mesh network interfaces. It is to be understood that the processor58controls the first CE device44to undertake present principles, including the other elements of the first CE device44described herein such as, e.g., controlling the display50to present images thereon and receiving input therefrom. Furthermore, note that the network interface56may be, for example, a wired or wireless modem or router, or other appropriate interface such as a wireless telephony transceiver, or Wi-Fi transceiver as mentioned above, etc.

Still further, note that in addition to the processor(s)58, the first CE device44may also include a graphics processing unit (GPU)55on a graphics card55A. The graphics processing unit55may be configured for, among other things, presenting AR and/or VR images on the display50.

In addition to the foregoing, the first CE device44may also include one or more input ports60such as, e.g., a HDMI port or a USB port to physically connect (e.g., using a wired connection) to another CE device and/or a headphone port to connect headphones to the first CE device44for presentation of audio from the first CE device44to a user through the headphones. The first CE device44may further include one or more tangible computer readable storage medium62such as disk-based or solid-state storage. Also in some embodiments, the first CE device44can include a position or location receiver such as but not limited to a cellphone and/or GPS receiver and/or altimeter64that is configured to, e.g., receive geographic position information from at least one satellite and/or cell tower, using triangulation, and provide the information to the CE device processor58and/or determine an altitude at which the first CE device44is disposed in conjunction with the CE device processor58. However, it is to be understood that that another suitable position receiver other than a cellphone and/or GPS receiver and/or altimeter may be used in accordance with present principles to, e.g., determine the location of the first CE device44in all three dimensions.

Continuing the description of the first CE device44, in some embodiments the first CE device44may include one or more cameras66that may be, e.g., a thermal imaging camera, an IR camera, a digital camera such as a webcam, and/or another type of camera integrated into the first CE device44and controllable by the CE device processor58to generate pictures/images and/or video in accordance with present principles. Also included on the first CE device44may be a Bluetooth transceiver68and other Near Field Communication (NFC) element70for communication with other devices using Bluetooth and/or NFC technology, respectively. An example NFC element can be a radio frequency identification (RFID) element.

Further still, the first CE device44may include one or more auxiliary sensors72(e.g., a motion sensor such as an accelerometer, gyroscope, cyclometer, or a magnetic sensor, an infrared (IR) sensor, an optical sensor, a speed and/or cadence sensor, a gesture sensor (e.g., for sensing gesture command), etc.) providing input to the CE device processor58. The first CE device44may include still other sensors such as, for example, one or more climate sensors74(e.g., barometers, humidity sensors, wind sensors, light sensors, temperature sensors, etc.) and/or one or more biometric sensors76providing input to the CE device processor58. In addition to the foregoing, it is noted that in some embodiments the first CE device44may also include an infrared (IR) transmitter and/or IR receiver and/or IR transceiver78such as an IR data association (IRDA) device. A battery (not shown) may be provided for powering the first CE device44. The CE device44may communicate with the AVD12through any of the above-described communication modes and related components.

The second CE device46may include some or all of the components shown for the CE device44. Either one or both CE devices may be powered by one or more batteries.

Now in reference to the afore-mentioned at least one server80, it includes at least one server processor82, at least one tangible computer readable storage medium84such as disk-based or solid-state storage. In an implementation, the medium84includes one or more solid state storage drives (SSDs). The server also includes at least one network interface86that allows for communication with the other devices ofFIG.1over the network22, and indeed may facilitate communication between servers and client devices in accordance with present principles. Note that the network interface86may be, e.g., a wired or wireless modem or router, Wi-Fi transceiver, or other appropriate interface such as a wireless telephony transceiver. The network interface86may be a remote direct memory access (RDMA) interface that directly connects the medium84to a network such as a so-called “fabric” without passing through the server processor82. The network may include an Ethernet network and/or fiber channel network and/or InfiniBand network. Typically, the server80includes multiple processors in multiple computers referred to as “blades” that may be arranged in a physical server “stack”.

Accordingly, in some embodiments the server80may be an Internet server or an entire “server farm”, and may include and perform “cloud” functions such that the devices of the system10may access a “cloud” environment via the server80in example embodiments for, e.g., domain adaptation as disclosed herein. Additionally, or alternatively, the server80may be implemented by one or more game consoles or other computers in the same room as the other devices shown inFIG.1or nearby.

FIG.2illustrates a system200the components of which may incorporate appropriate components shown inFIG.1. A computer simulation controller202such as a PlayStation® controller, Xbox® controller, or other controller may include a touchpad204that can receive touch signals from a hand206and communicate via wired and/or wireless paths208with a computer simulation console210and/or a display device212such as an Internet-enabled TV. As explained further below, the user can manipulate the touchpad204to generate alpha-numeric characters214for presentation on the display device212either through direct communication of signals with the display device or through the simulation console210. More specifically, by manipulating the touchpad204, a user can move a screen cursor over a letter on a virtual keyboard216presented on the display device212to enter the alpha-numeric characters214. The virtual keyboard216may have, without limitation, a QWERTY layout.

As shown schematically inFIG.3, present principles contemplate two types of touch, namely, a “soft” press300(using a soft pressure on the touchpad or a hover over the touchpad with zero pressure), in which a screen cursor on the display device212is moved to desired locations on the virtual keyboard216without sending location data (i.e., a signal indicating selection of any particular virtual key) to the display device, and a “hard” press302of greater pressure than a soft press, in response to which a screen cursor on the display device212may be moved and location data sent to the display device to indicate selection of a virtual key. In this way, a user can look away from the touchpad204and view the virtual keyboard216while moving his or her finger across the touchpad to move a visible screen cursor to a desired letter on the virtual keyboard, and then exert a hard press to select that letter. Note that an individual “next” letter may not be presented on the display, but rather the next “most possible word” may be displayed after a user has finished a “swipe”. The “hottest” key (based on the heatmap) may be highlighted on the virtual keyboard as well as the trace. In addition, a “swipe” is defined as a continuous hard-press which forms a trace.

FIG.4illustrates example logic with the above description in mind. The logic may be executed by one or more of a processor in the simulation controller202, a processor in the simulation console210, and a processor in the display device212.

Commencing at state400it is determined whether a press of the touchpad204has been received. This may be done by determining whether signals from one or more proximity sensors associated with the touchpad204indicate a hover of a finger adjacent the touchpad204and/or by determining whether signals from one or more pressure sensors associated with the touchpad204indicate a pressure of at least a first threshold pressure.

When it is determined that a touch has been received, the logic proceeds to state402to determine whether the touch is a soft press or hard press as indicated by, e.g., signals from a pressure sensor associated with the touchpad204indicating a touch of at least a threshold pressure, which is typically set to be greater than any threshold pressure used at state400. If the touch does not satisfy the threshold, the logic moves to block404to return a soft press. In some implementations the logic may proceed to state406to determine whether the soft press is the first soft press within, e.g., a threshold of period, for example within the last five minutes, and if so the logic can move to block408to enlarge an image of the virtual keyboard216on the display device212. In any case, from state406if the test there is negative or from block408, the logic moves to block410to move the screen cursor without sending press location information.

On the other hand, if the test at state402determines that a hard press is received, such is returned at block412, and the screen may be moved according to the touch with location information being sent as well indicating the location of the virtual keyboard the user has selected by means of the hard press on the touchpad204of the simulation controller200.

FIG.5illustrates a combination of hardware and software blocks alluded to above.

One or more proximity and/or pressure sensors500are provided in the touchpad204to output signals representing soft presses502and hard presses504. The soft presses502establish finger focus points506. The hard presses504establish points on the touchpad as detected by the sensor(s)500. A soft-press represents a cursor focus point, while “points by sensor” means “continuous points sending by the sensor”.

At510a heatmap algorithm, discussed further below in reference toFIGS.6-8, is accessed to output a sequence of letters512according to the hard presses504. The sequence of letters512is input along with a dictionary514to a reduction block516that reduces the list of candidates that might possibly form either a correction to or a completion of the sequence of letters512. The dictionary514is essentially a dictionary and/or thesaurus of sequences of letters that can be used to correct a mis-typed word, e.g., the dictionary514may correlate “thw” to “the” to return the word “the” in response to input of “thw”.

The reduced list of candidates516is provided to a module518that outputs a predicted network or words for presentation on the screen, which a user can then select to complete his or her desired input without typing every letter of the predicted word or words. The module518may be established by one or more neural networks (NN) as described further below. To produce a predicted word or words, the module518may receive input from a contextual user block520, which provides previous word strings employed by the user with the current input inferred to possibly be a repeat of a prior input, e.g., “do you” may have been followed multiple times in prior inputs by “know what I mean”, and this information can be input to help train and execute the module518.

Moreover, similar training/execution aids may be input to the module518as shown at the right ofFIG.5. Specifically, queries and chat data522from other computer gamers may be input to a character-based NN such as a bidirectional long short-term memory (BILSTM)524to learn patterns of common input strings for provision to a machine learning character sequence model526. This model526may be input to or accessed by the module518in rendering a next predicted word or words.

FIGS.6-8Billustrate employment of the heatmap algorithm510inFIG.5. Basically, the “path” or “connected points” of the finger “swipe” (hard-press) and the probabilities of each letter are “discounted and accumulated” at certain time interval along the swipe. At each time interval, the letter with the highest probability is extracted, which may also have to pass a certain threshold to add to the sequence as developed further below.

InFIGS.6-8B, it is to be understood that only the first four letters in the top left corner of a QWERTY keyboard (i.e., Q, next to which is “W”, and below which from left to right are “A” and “S”) are shown for clarity of disclosure, as but one example of a possible virtual keyboard layout for the virtual keyboard216. In the example heatmap510illustrated, each area of the heatmap for a particular letter is divided into a three-by-three grid for nine divisions (illustrated as geometric squares) total, with the center division600for a particular letter indicating that the probability of that letter being desired when a cursor is in the center area being 1. In contrast, the heatmap510indicates probabilities less than one but greater than zero in the border divisions602that surround the center division600of a letter, with the probabilities being associated with the letter of the center division600and the letter(s) immediately adjacent the border divisions602(or, in the case of a border division that is not adjacent another letter, only a probability less than one for the letter of the center division).

As shown inFIG.7at700, a soft press is used to locate the starting letter of an intended input. Then, as shown at800inFIG.8, a hard press is used to indicate selection of the starting letter, in the example shown, “Q”. This causes the collection of data that “Q” is selected with a probability of one and that surrounding letters (in the example shown, “W”, “A”, and “S”) are not selected, i.e., have a probability of zero.

FIGS.8A and8Billustrate the results of an ensuing swipe. InFIG.8Aa swipe is shown at802from the location starting inFIG.8to the location804indicated by the image of the hand. Here, the user has moved his finger toward the letter “A”. This causes new heatmap statistics to be aggregated according to the path of the swipe over the border divisions602using the algorithm shown inFIG.8A. Because the probability of “Q” is higher than the probabilities of “W” (which is zero), “A” (which is 0.3), and “S” (which is zero), the sequence returns “Q”.

FIG.8Bshows at806that the swipe has been continued to the location808shown by the image of the hand. This causes further heatmap statistics to be aggregated according to the path of the swipe over the border divisions602using the algorithm shown inFIG.8B. Because the probability of “A” is higher than the probabilities of “W” (which is zero), “Q” (which is 0.3), and “S” (which is zero), the sequence returns “A” to be appended after “Q” was returned inFIG.8A, resulting in a sequence “QA”.

Thus, it may now be appreciated that the “path” or “connected points” of the finger “swipe” (hard-press) is tracked and the probabilities of each letter are discounted and accumulated at certain time intervals along the swipe. At each time interval, the letter with the highest probability is extracted, in some embodiments provided the probability of the letter satisfies a threshold probability (e.g., of 0.4) to be added to the sequence.

FIG.9illustrates an example NN architecture that may be used in any of the NN-based modules of, e.g.,FIG.5. A network900of NN may receive input letters902with probabilities904from the heatmap to output time-distributed predicted letters906with associated probabilities908. In the example shown, each letter902may be input to a respective recurrent NN (RNN) such as a sequence of long short-term memory (LSTM)910as shown. An LSTM910as shown at the right inFIG.9may include an input gate912, a forget gate914, and an output gate916, all of which may execute a sigmoid function as indicated by the Greek letter a inFIG.9. The input gate912controls the extent to which a new value flows into the cell, the forget gate914controls the extent to which a value remains in the cell and the output gate916controls the extent to which the value in the cell is used to compute the output activation of the LSTM unit.

The current value xibeing input and the hidden state ht-1from the previous iteration are input to all three gates as shown. The output of the sigmoid function of the input gate912may be combined with a hyperbolic tangent function918at a first combine operator920, which may be an element-wise product. The output of the first combine operator920is combined, as by summing if desired, with the output of s second combine operator922at a third combine operator924. The output of the third combine operator924may be fed back to the second combine operator922for combining with the output of the forget gate914. Further, the output of the third combine operator924may be operated on if desired by a hyperbolic tangent function926and then combined at a fourth combine operator928with the output of the output gate916to render a hidden state vector930for use in the succeeding iteration.

FIGS.10-12illustrate a sequence of the use of the network900to generate predicted text. The lower row of letters1000represents input received from hard presses on keys of the virtual keyboard216and/or from selection of previously predicted letters and/or words. These are input to the trained network900. Using probabilities correlated with letters from the heatmap as illustrated at1002, a next predicted letter1004is generated and fed back to the model. The sequence shown inFIGS.10-12generated predicted letters for an initial input of “play” that results in the word “PlayStation”.

FIG.13is a flow chart of example logic consistent with present principles. The NN system(s) described herein are trained at block1300. Moving to block1302, a hard press is received on the touchpad and a letter established based thereon at block1304using the heatmap if desired. The letter is input to the NN system at block1306, which outputs a predicted letter or words or string of words at block1308. The predicted letters/words are presented on screen at block1310.

If a user does not accept the predictions at state1312, they may be removed from presentation at state1314. Otherwise, accepted predictions are confirmed at block1316and presented in sequence after the letters established by the hard press.

Present principles may be used in all possible deep learning-based methods for image, video and audio data processing, among others.

As may be appreciated from the foregoing detailed description, present principles thus improve the adaptation and training of neural networks through the technological solutions described herein.

It will be appreciated that whilst present principals have been described with reference to some example embodiments, these are not intended to be limiting, and that various alternative arrangements may be used to implement the subject matter claimed herein.