Auto-complete methods for spoken complete value entries

A voice-enabled system (fig.4) for guiding a user through a particular task prompts a user to speak spoken complete value entry 120 (eg. workflow steps in a task, or credit card/vehicle identification numbers), receives a spoken subset of the complete long value 130 (eg. the first few digits of the...

Full description

Saved in:
Bibliographic Details
Main Authors Russell Evans, Lauren Meyer, Mark Koenig, Alexander Nikolaus Miracna, Bernard Kriley, Kurt Charles Miller, Craig George, Matthew Nichols, Brian Manuel, Luke Sadecky
Format Patent
LanguageEnglish
Published 10.05.2017
Subjects
ACOUSTICS
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
MUSICAL INSTRUMENTS
PHYSICS
SPEECH ANALYSIS OR SYNTHESIS
SPEECH OR AUDIO CODING OR DECODING
SPEECH OR VOICE PROCESSING
SPEECH RECOGNITION
Online AccessGet full text

Cover

More Information
Summary:A voice-enabled system (fig.4) for guiding a user through a particular task prompts a user to speak spoken complete value entry 120 (eg. workflow steps in a task, or credit card/vehicle identification numbers), receives a spoken subset of the complete long value 130 (eg. the first few digits of the number, the minimum number of such characters being pre-selected), compares this subset with all the possible complete entries in a suggestion list (received or compiled previously, 105), and predicts the intended complete value in order to automatically complete the entry 150. The user may be alerted should no list entry match the spoken subset, while correct predictions can be confirmed 160 or rejected by the user via eg. a graphical user interface. The mobile device (16, fig. 1) may notify a server (12) of the confirmation.
Bibliography:Application Number: GB20160013949