Autors: Pleshkova, S. G., Bekyarski, A. B.
Title: Development of Models for Speech Recognition and Natural Language Understanding Using IoT Modules with Parallel Architectures
Keywords: Speech recognition models, Parallel IoT architecture, Natura

Abstract: Speech recognition is one of the main methods by which artificial intelligence models a person's ability to perceive and communicate through speech. In order to achieve in speech recognition the human ability to perceive and understand speech, it is necessary to improve the already existing and achieved in practical use methods and algorithms for speech recognition and natural languages understanding. This can be done by creating pre-designed models with established accuracy to be used in various specific practical implementations of speech recognition applications. The purpose of this article is to use such predefined models and embed them in modules of Internet of Things (IoT), which have a parallel architecture and would allow real-time speech recognition.

References

  1. NVIDIA Cloud Computing. https://www.nvidia.com/en-us/data-center/gpu-cloud-computing/
  2. Jetson Nano Developer Kit, https://developer.nvidia.com/jetson-nano-developer-kit
  3. [1] A. Graves and J. Schmidhuber,, 2005, Framewise phoneme classification with bidirectional lstm and other neural network architectures, Neural Networks, Volume vol. 18,, pp. pp. 602–610

Issue

CEMA 2022, 2022, Greece, ISBN ISSN 1314-2100

Full text of the publication

Вид: пленарен доклад в международен форум, индексирана в Scopus