Patent · US Active

Method for speech processing and speech processing device

US10861463B2 · kind B2 · utility

0Cited by
8References
21Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJan 9, 2018
Grant dateDec 8, 2020
Priority date
Expiry dateJun 30, 2038

Classification

  • Technology area (CPC H)Electricity
  • CPC primaryH04M2203/306
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

During telephone calls in a public space, users may be reluctant to provide private or secret information, due to the risk of eavesdropping. A hands-free solution for entering secret information to electronic speech communication devices is based on speech processing. A method for speech processing a voice input data stream comprises steps of scanning the voice input data stream and detecting a spoken delimiter therein, determining a predefined audio sample corresponding to the detected spoken delimiter, inserting the determined predefined audio sample into the voice input data stream at the spoken delimiter, wherein a substituted voice data stream is obtained and wherein speech portions of the voice input data stream at least before the spoken delimiter remain in the substituted voice data stream, and providing the substituted voice data stream for output towards a recipient.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.