Patents by Inventor Ondrej Bojar

Ondrej Bojar has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Computer-Implemented Method Of Real Time Speech Translation And A Computer System For Carrying Out The Method
    Publication number: 20230306207
    Abstract: A computer-implemented method of real time speech translation wherein at least a source speech and a human interpretation of the source speech are transcribed using an automatic speech recognition system and machine translated into a common language. A best source of data is then selected repeatedly and data from the best source are machine translated into at least one another language.
    Type: Application
    Filed: March 22, 2022
    Publication date: September 28, 2023
    Applicant: Charles University, Faculty Of Mathematics And Physics
    Inventors: Ondrej Bojar, Dominik Machacek
  • Computer-implemented method of creating a translation model for low resource language pairs and a machine translation system using this translation model
    Patent number: 11037028
    Abstract: A computer-implemented method for creating a translation model for low resource language pairs and applicable on noisy inputs utilizing several approaches: choosing particular input corpora covering in-domain noisy and clean texts as well as unrelated but larger general parallel texts, performing several chosen methods of creating synthetic parallel corpora and filtering, pre-processing, deduplicating and concatenating training corpora.
    Type: Grant
    Filed: December 31, 2018
    Date of Patent: June 15, 2021
    Assignee: Charles University Faculty of Mathematics and Physics
    Inventors: Ondrej Bojar, Roman Sudarikov
  • A Computer-Implemented Method of Creating a Translation Model for Low Resource Language Pairs and a Machine Translation System using this Translation Model
    Publication number: 20200210772
    Abstract: A computer-implemented method for creating a translation model for low resource language pairs and applicable on noisy inputs utilizing several approaches: choosing particular input corpora covering in-domain noisy and clean texts as well as unrelated but larger general parallel texts, performing several chosen methods of creating synthetic parallel corpora and filtering, pre-processing, deduplicating and concatenating training corpora.
    Type: Application
    Filed: December 31, 2018
    Publication date: July 2, 2020
    Applicant: Charles University Faculty of Mathematics and Physics
    Inventors: Ondrej Bojar, Roman Sudarikov