# US 2014/0198047 A1 · Reducing error rates for touch based keyboards > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2014/0198047 A1 - Document ID: US20140198047A1 - Status: Published application - Kind: A1 - Application number: 13/830,674 - Filed: 2013-03-14 - Published: 2014-07-17 - Earliest claimed: 2013-01-14 - Assignees: Nuance Communications, Inc. - Applicants: Nuance Communications, Inc. - Inventors: Erland Unruh; David Kay - Web page: https://patentgrep.com/patent/US20140198047A1 ## CPC - G06F 3/0412: {Digitisers structurally integrated in a display} - G06F 3/0237: {using prediction or retrieval techniques} - G06F 3/04886: by partitioning the display area of the touch-screen or the surface of the digitising tablet into independently controllable areas, e.g. virtual keyboards or menus - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract The present technology provides systems and methods for reducing error rates to data input to a keyboard, such as a touch screen keyboard. In one example, an input bias model dynamically changes the keyboard functionality such that the keyboard will not necessarily produce the same result for an identical tap coordinate. Rather, the keyboard functionality is adapted to account for key offset bias that occurs when the user has a tendency to select a tap coordinate that would otherwise return an unintended key. Additionally, the present technology provides a language feedback model that may provide a probability for a next tap coordinate and may augment the key corresponding to the most probable next tap coordinate, thereby allowing the user to more easily select the correct key. Further details are provided herein. ## Claims ### Claim 1 A method for reducing error rates associated with key input to a keyboard, the method comprising: detecting at least one keyboard event, wherein the detected keyboard event includes a set of coordinates indicated by a touch of a user on a keyboard, and wherein each key on the keyboard is represented by coordinates defining an area on the keyboard for that key; determining an intended input key based on the detected keyboard event, wherein determining the intended input key includes: determining multiple candidate keys near to the set of coordinates indicated by the touch on the keyboard; calculating a probability for whether each of the multiple candidate keys was the intended input key; and, updating the defined area for a key on the keyboard based on the determined, intended input key. ### Claim 2 (depends on 1) The method of claim 1, further comprising: determining a selected input key; comparing the selected input key to the intended input key; and if the selected input key does not match the intended input key: identifying a type of detected keyboard event; retrieving a weighting factor for the identified keyboard event type; increasing a weighting of the selected input key at the set of coordinates according to the retrieved weighting factor; or decreasing the weighting of the selected input key at the set of coordinates according to the retrieved weighting factor. ### Claim 3 (depends on 1) The method of claim 1, wherein the keyboard is a QWERTY touch-screen keyboard provided by a touchscreen on a mobile device, and wherein the type of keyboard event includes a delete event, a manual word selection event, or an automatic word selection event. ### Claim 4 (depends on 1) The method of claim 1, wherein determining the intended key and updating the defined area on the keyboard is different based on at least two of: a type of keyboard, a language, a particular user, and a particular device. ### Claim 5 (depends on 1) The method of claim 1, wherein the keyboard is a virtual keyboard, wherein each key on the virtual keyboard is further defined by center coordinates of a probability distribution for intended selection of that key, wherein the center coordinates correspond to the highest probability of the probability distribution, and wherein updating the defined area includes moving the center coordinates to a new location with respect to the virtual keyboard. ### Claim 6 (depends on 1) The method of claim 1, wherein determining the intended key includes: comparing multiple, previously input characters to a language model, wherein the language model includes at least one dictionary of words; and, determining probabilities of multiple next most likely characters based on the language model. ### Claim 7 (depends on 1) The method of claim 1, wherein determining the intended key includes: employing a static probability model for each key of the keyboard, wherein at least a portion of coordinates defining the area on the keyboard for each key cannot be changed by the method. ### Claim 8 (canceled) ### Claim 9 (canceled) ### Claim 10 (canceled) ### Claim 11 (canceled) ### Claim 12 (canceled) ### Claim 13 (canceled) ### Claim 14 (canceled) ### Claim 15 A system for improving input to a virtual or touch keyboard, the system comprising: at least one processor; at least one data storage device storing instructions performed by the processor; a keyboard coupled to the processor; a first set of instructions for detecting at least one keyboard event, wherein the detected keyboard event includes a set of coordinates indicated by a touch of a user on the keyboard, and wherein each key on the keyboard is defined by coordinates defining an area on the keyboard for that key; means for determining an input key based on the detected keyboard event, wherein determining the input key includes: determining multiple candidate keys; calculating a probability for the input key; and, means for updating the defined area on the keyboard based on the determined input key. ### Claim 16 (depends on 15) The system of claim 15, further comprising: means for determining a selected input key based on a keyboard landscape; means for comparing the selected input key to an intended input key; and if the selected input key does not match the intended input key: means for identifying a type of detected keyboard event; means for retrieving a weighting factor for the identified keyboard event type; means for increasing a weighting of the selected input key at the set of coordinates according to the retrieved weighting factor; or means for decreasing the weighting of the selected input key at the set of coordinates according to the retrieved weighting factor. ### Claim 17 (depends on 15) The system of claim 15, wherein the system is portable data processing device, wherein the keyboard is a virtual keyboard provided by a touchscreen that displays a QWERTY keyboard, wherein the type of keyboard event includes a delete event, a manual word selection event, or an automatic word selection event. ### Claim 18 (depends on 15) The system of claim 15, wherein determining the input key and updating the defined area on the keyboard is different based on at least two of: a type of keyboard, a language, a user, and a device. ### Claim 19 (depends on 15) The system of claim 15, wherein each key on the keyboard is further defined by center coordinates for a probability distribution, and wherein updating the defined area includes moving the center coordinates to a new location with respect to the keyboard. ### Claim 20 (depends on 15) The system of claim 15, wherein determining the input key includes: comparing multiple, previously input characters to a language model, wherein the language model includes at least one dictionary of words; and, determining probabilities of next most likely characters based on the language model. ### Claim 21 A tangible computer-readable medium storing instructions that, when executed by at least one processor, reduce error rates associated with key input to a keyboard, comprising: detecting at least one keyboard event, wherein the detected keyboard event includes a set of coordinates that map to a keyboard location that is tapped by a user, and wherein each key on the keyboard corresponds to a set of coordinates outlining an area on the keyboard for that key; determining an intended input key based on the detected keyboard event, wherein determining the intended input key includes: determining multiple candidate keys in the vicinity of the set of coordinates that map to the tapped keyboard location; calculating a likelihood for whether each of the multiple candidate keys was the intended input key; and, updating the outlined area for a key on the keyboard based on the determined, intended input key. ### Claim 22 (depends on 21) The tangible computer-readable medium of claim 21, further comprising: determining a selected key; comparing the selected key to the intended key; and if the selected key does not match the intended key: identifying a category corresponding to the detected keyboard event; retrieving a weighting factor for the identified keyboard event category; increasing a weighting of the selected key at the set of coordinates in accordance with the retrieved weighting factor; or decreasing the weighting of the selected key at the set of coordinates in accordance with the retrieved weighting factor. ### Claim 23 (depends on 21) The tangible computer-readable medium of claim 21, wherein the keyboard is a QWERTY touch-screen keyboard provided by a touchscreen on a portable device, and wherein the category of keyboard event includes a delete event, a manual word selection event, or an automatic word selection event. ### Claim 24 (depends on 21) The tangible computer-readable medium of claim 21, wherein determining the intended key and updating the outlined area on the keyboard is different based on at least two of: a type of keyboard, a language, a particular user, and a particular device. ### Claim 25 (depends on 21) The tangible computer-readable medium of claim 21, wherein the keyboard is a virtual keyboard, wherein each key on the virtual keyboard is further outlined by center coordinates of a probability distribution for intended selection of that key, wherein the center coordinates correspond to the highest probability of the probability distribution, and wherein updating the outlined area includes relocating the center coordinates to a different location with respect to the virtual keyboard. ### Claim 26 (depends on 21) The tangible computer-readable medium of claim 21, wherein determining the intended key includes: comparing multiple, previously input characters to a language model, wherein the language model includes at least one dictionary of words; and, determining probabilities of multiple next most likely characters based on the language model. ### Claim 27 (depends on 21) The tangible computer-readable medium of claim 21, wherein determining the intended key includes: using a static probability model for each key of the keyboard, wherein at least a portion of coordinates outlining the area on the keyboard for each key cannot be changed by the method.