US 12,347,426 B2
Computer-based systems utilizing speech pattern recognition machine learning models and methods of use thereof
Stephen L. Jordan, Burke, VA (US); and Sean Easter, Mahtomedi, MN (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Oct. 17, 2022, as Appl. No. 17/967,736.
Prior Publication US 2024/0127806 A1, Apr. 18, 2024
Int. Cl. G10L 15/00 (2013.01); G06F 40/166 (2020.01); G06F 40/284 (2020.01); G06N 3/044 (2023.01); G06N 3/08 (2023.01); G10L 15/06 (2013.01); G10L 15/16 (2006.01); G10L 15/197 (2013.01); G10L 15/22 (2006.01)
CPC G10L 15/197 (2013.01) [G06F 40/166 (2020.01); G06F 40/284 (2020.01); G06N 3/044 (2023.01); G06N 3/08 (2013.01); G10L 15/063 (2013.01); G10L 15/16 (2013.01); G10L 15/22 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
receiving, by a processor, at least one transcribed speech file of speech of at least one person;
identifying, by the processor, a plurality of utterance-delimited multiword units in the at least one transcribed speech file;
wherein each utterance-delimited multiword unit corresponds to an identified utterance in the speech delimited by pauses in the speech of the at least one person;
generating, by the processor, a plurality of utterance-based data vectors corresponding to the plurality of utterance-delimited multiword units by assigning a numerical data token to each word in each utterance-delimited multiword unit;
wherein each utterance-based data vector comprises a sequence of numerical data tokens;
determining, by the processor, at least one utterance-delimited multiword unit from the plurality of utterance-delimited multiword units having question-indicative word patterns by inputting the plurality of utterance-based data vectors into at least one machine learning model, trained to output a probability that a word, groups of words, or both in each of the plurality of utterance-delimited multiword units comprises the question-indicative word patterns; and
identifying, by the processor, at least one question in the at least one transcribed speech file of the speech of the at least one person based at least in part on:
the probability and
the question-indicative word patterns identified in the at least one utterance-delimited multiword unit.
 
11. A system, comprising:
a memory configured to store computer code; and
a processor configured to execute the computer code stored in the memory that causes the processor to:
receive at least one transcribed speech file of speech of at least one person;
identify a plurality of utterance-delimited multiword units in the at least one transcribed speech file;
wherein each utterance-delimited multiword unit corresponds to an identified utterance in the speech delimited by pauses in the speech of the at least one person;
generate a plurality of utterance-based data vectors corresponding to the plurality of utterance-delimited multiword units by assigning a numerical data token to each word in each utterance-delimited multiword unit;
wherein each utterance-based data vector comprises a sequence of numerical data tokens;
determine that at least one utterance-delimited multiword unit from the plurality of utterance-delimited multiword units having question-indicative word patterns by inputting the plurality of utterance-based data vectors into at least one machine learning model, trained to output a probability that a word, groups of words, or both in each of the plurality of utterance-delimited multiword units comprises the question-indicative word patterns; and
identify, based on the probability, at least one question in the at least one transcribed speech file of the speech of the at least one person based at least in part on:
the probability, and
the question-indicative word patterns identified in the at least one utterance-delimited multiword unit.