During operation of an electronic device such as a cellular telephone with a touch screen display or other electronic equipment, a voice recognition engine may gather data on spoken words. Data on the spoken words that are recognized may be maintained in a spoken word database maintained by an input processor with an autocorrection engine. A user may supply text input that contains mistyped words to the electronic device using the touch screen or a keyboard. The input processor may use the autocorrection engine to automatically replace mistyped words with corrected versions of the mistyped words. The corrected words may be displayed in real time as the user supplies the text input. The autocorrection engine may make word correction decisions based at least partly on information in the spoken word database.