Publication detail

Human-Machine Interface for Mobile Robot Based on an Enhancement Speech Recognition

MAŠEK, P. RŮŽIČKA, M.

English title

Human-Machine Interface for Mobile Robot Based on an Enhancement Speech Recognition

Type

Paper in proceedings (conference paper)

Language

en

Original abstract

The paper deals with human-machine interface for mobile robot based on enhancement speech recognition system. The speech recognition system can be based on both the various commonly used commercial or open sources. Instead of developing a complete solution from scratch, the third-party system was used. Despite the fact, such solution shows useful results, there are some limitations of uses this kind of speech recognition engine for communication with autonomous robot (e.g. low success rate in recognition of specific words, wrong interpretation in noisy environment, etc.). We present a solution in which the well known Bayesian approach is used for enhance the results and there are described experimental result obtained on autonomous mobile robot in real environment.

Keywords in English

Speech recognition, Bayesian filter, Probabilistic learning, Human-machine interface, Mobile robot.

Released

2014-06-25

Publisher

Brno University of Technology, Faculty of Mechanical Engineering, Institute of Automation and Computer Science

Location

Brno, Czech Republic

ISBN

978-80-214-4984-8

ISSN

1803-3814

Book

MENDEL 2014, 20th International Conference on Soft Computing.

Volume

2014

Number

20

Pages from–to

249–252

Pages count

4

BIBTEX


@inproceedings{BUT108363,
  author="Petr {Mašek} and Michal {Růžička}",
  title="Human-Machine Interface for Mobile Robot Based on an Enhancement Speech Recognition",
  booktitle="MENDEL 2014, 20th International Conference on Soft Computing.",
  year="2014",
  series="1",
  journal="Mendel Journal series",
  volume="2014",
  number="20",
  pages="249--252",
  publisher="Brno University of Technology, Faculty of Mechanical Engineering, Institute of Automation and Computer Science",
  address="Brno, Czech Republic",
  isbn="978-80-214-4984-8",
  issn="1803-3814"
}