Approach to reducing the response time of a speech interface
AssigneeInternational Business Machines Corporation
InventorsGakuto Kurata, Tohru Nagano
Life of this patent
Estimating the term…
Abstract
A method for reducing response time in a speech interface including constructing a partially completed word sequence from a partially received utterance from a speaker received by an audio sensor, modeling a remainder portion using a processor based on a rich predictive model to predict the remainder portion, and responding to the partially completed word sequence and the predicted remainder portion using a natural language vocalization generator with a vocalization, wherein the vocalization is prepared before a complete utterance is received from the speaker and conveyed to the speaker by an audio transducer.
Claims20
A computer implemented method for reducing response time in a speech interface comprising:
- receiving a series of sounds at an audio sensor;
- parsing the received sounds with an automatic speech recognition detector to identify one or more words of a partially received utterance;
- predicting words for which sounds have not yet been received by the audio sensor by determining an intended semantic or context from the partially received utterance and the purpose and function of the speech interface;
- and
- modeling the probability of words forming a remainder portion of the partially received utterance to produce a predicted remainder portion.
A non-transitory computer readable storage medium comprising a computer readable program for reducing response time in a speech interface, wherein the computer readable program when executed on a computer causes the computer to perform the steps of:
- parsing a series of sounds received at an audio sensor with an automatic speech recognition detector to identify one or more words of a partially received utterance;
- predicting words for which sounds have not yet been received by the audio sensor by determining an intended semantic or context from the partially received utterance and the purpose and function of the speech interface;
- and
- modeling the probability of words forming a remainder portion of the partially received utterance to produce a predicted remainder portion.
A system for reducing response time in a speech interface comprising:
- an automatic speech recognition detector including a processor configured to parse a series of sounds received at an audio sensor to identify one or more words of a partially received utterance;
- and
- a word predictor configured to predict words for which sounds have not yet been received by the audio sensor by determining an intended semantic or context from the partially received utterance and the purpose and function of the speech interface, and modeling the probability of words forming a remainder portion of the partially received utterance to produce a predicted remainder portion.
Dependencies are inferred from the claim text.