Text input using key and gesture information

摘要

Text input may be identified from a combination of key stroke information and handwriting gesture information. In one example, a touch screen displays an on-screen keyboard. A user then draws a character on the keyboard over the key that represents the character. In this way, two types of information are provided that identify the character that the user intended to enter: the particular drawing gesture that the user provided, and the location on the keyboard at which the user drew that gesture. These two pieces of information may be used, in combination, to determine which character the user intended to enter, and may help to interpret the input accurately in the case where either the gesture or key information, individually, would have been ambiguous.

主权项

1. One or more device-readable storage media that store executable instructions to receive text input, wherein the executable instructions, when executed on a device, cause the device to perform acts comprising:
receiving a gesture drawn by a user on a touch screen on which an on-screen keyboard is being displayed at a time at which the gesture is drawn; identifying a character represented by the gesture; said gesture comprising a shape of said character; identifying a key over which the gesture is drawn; determining that the user entered the character based on the gesture and also based on the key; identifying one or more characters that correspond to the gesture, wherein each of the one or more characters is associated with a probability that it corresponds to the gesture; identifying one or more keys over which the gesture is drawn, wherein each of the one or more keys is associated with a probability that it is the key that the user intended to indicate; using a weighted average of probabilities associated with the characters and probabilities associated with the keys to determine which character the user entered; and storing, communicating, or displaying the character.