Environment sensitive predictive text entry
AssigneeVerizon Patent and Licensing Inc.
InventorsDonald G. Archer, Andrien J. Wang, Aravind Perumandla
Life of this patent
Estimating the term…
Abstract
Environmental factors may be used in a predictive text system provided in a device. The device may receive one or more characters entered by a user and determine, based on the one or more characters, words that are predicted to be words being entered by the user of the device, where the words are determined using grammar-based predictive techniques. The device may determine confidence scores, corresponding to the words. The device may refine the scores based on environmental data that includes data that describes an environment associated with the user of the device. The device may select, based on the refined plurality of scores, a subset of the plurality of words and output the subset of the plurality of words.
Claims21
A method, implemented by a mobile device, the method comprising:
- receiving, by the mobile device, one or more characters entered by a user of the mobile device;
- determining, by the mobile device and based on the one or more characters, a plurality of words that are predicted to be words being entered by the user of the mobile device, the plurality of words being determined using grammar-based predictive techniques;
- determining, by the mobile device, a plurality of scores, corresponding to the plurality of words, each of the plurality of scores representing an estimated confidence that the corresponding one of the plurality of words is the word that is being entered by the user;
- refining, by the mobile device, the plurality of scores based on environmental data that includes data that describes an environment associated with the user of the mobile device, the refining of the plurality of scores including refining the plurality of scores based on rules that relate parameters of the environmental data to score modifications;
- selecting, by the mobile device and based on the refined plurality of scores, a subset of the plurality of words;
- and
- outputting, by the mobile device, the subset of the plurality of words, to a display of the mobile device, as predicted text that is selectable by the user to complete a word or phrase corresponding to the one or more characters entered by the user.
A device comprising:
- a processor to:
- receive one or more characters entered by a user of the device;
- determine, based on the one or more characters, a plurality of words that are predicted to be words being entered by the user of the device, the plurality of words being determined using grammar-based predictive techniques;
- determine a plurality of scores, corresponding to the plurality of words, each of the plurality of scores representing an estimated confidence that the corresponding one of the plurality of words is the word that is being entered by the user;
- refine the plurality of scores based on environmental data that includes data that describes an environment associated with the user of the device, the refining of the plurality of scores including refining the plurality of scores based on rules that relate parameters of the environmental data to score modifications;
- select, based on the refined plurality of scores, a subset of the plurality of words;
- and
- output, the subset of the plurality of words, to a display of the device, as predicted text that is selectable by the user to complete a word or phrase corresponding to the one or more characters entered by the user.
A non-transitory computer-readable medium, comprising:
- one or more instructions that, when executed by one or more processors of a device, cause the one or more processors to:
- receive one or more characters entered by a user;
- determine, based on the one or more characters, a plurality of words that are predicted to be words being entered by the user, the plurality of words being determined using grammar-based predictive techniques;
- determine a plurality of scores, corresponding to the plurality of words, each of the plurality of scores representing an estimated confidence that the corresponding one of the plurality of words is the word that is being entered by the user;
- refine the plurality of scores based on environmental data that includes data that describes an environment associated with the user, the refining of the plurality of scores including successively refining the plurality of scores by a plurality of environment modules, where each environment module, of the plurality of environment modules, refines the plurality of scores based on different environmental data;
- select, based on the refined plurality of scores, a subset of the plurality of words;
- and
- output, the subset of the plurality of words, as predicted text that is selectable by the user to complete a word or phrase corresponding to the one or more characters entered by the user.
Dependencies are inferred from the claim text.