US 12,243,515 B2
Speech recognition using neural networks
Andrew W. Senior, New York, NY (US); and Ignacio L. Moreno, New York, NY (US)
Assigned to Google LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on Mar. 2, 2023, as Appl. No. 18/177,717.
Application 18/177,717 is a continuation of application No. 17/154,376, filed on Jan. 21, 2021, granted, now 11,620,991.
Application 17/154,376 is a continuation of application No. 16/573,232, filed on Sep. 17, 2019, granted, now 10,930,271, issued on Feb. 23, 2021.
Application 16/573,232 is a continuation of application No. 13/955,483, filed on Jul. 31, 2013, granted, now 10,438,581, issued on Oct. 8, 2019.
Prior Publication US 2023/0206909 A1, Jun. 29, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G10L 15/02 (2006.01); G06N 3/02 (2006.01); G10L 15/16 (2006.01)
CPC G10L 15/16 (2013.01) [G06N 3/02 (2013.01); G10L 15/02 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method when executed on data processing hardware causes the data processing hardware to perform operations comprising:
while a user is speaking a current utterance:
obtaining feature vectors indicative of audio characteristics of corresponding portions of the current utterance that have been spoken by the user;
for each predetermined duration of new audio received that characterizes a corresponding portion of the current utterance, calculating a corresponding i-vector;
providing, as input to a neural network acoustic model, the feature vectors and the corresponding i-vector calculated for each predetermined duration of new audio received; and
based on the feature vectors and the corresponding i-vector calculated for each predetermined duration of new audio received and provided as input to the neural network acoustic model, determining, as output from an output layer of the neural network acoustic model, a posterior probability distribution of possible speech units representing each feature vector.