Header menu link for other important links
X
Implementation of machine learning applications on a fixed-point DSP
Published in Institute of Electrical and Electronics Engineers Inc.
2015
Volume: 2015-June
   
Issue: June
Pages: 1458 - 1463
Abstract
In this paper, we discuss efficient implementation of machine learning algorithms on DSPs. Specifically, we implement OCR and speech recognition on DSP and show how they can be optimized using fixed point routines. We illustrate the optimal usage of DSP resources like MAC units, shifters and software pipelining through assembly code structuring which massively reduces the MIPS consumed by the processor. We also describe how floating point overheads can be reduced by equivalent fixed point routines for real time implementations. Though the Blackfin-533 DSP is chosen for this illustration, the ideas presented here apply to other fixed point DSPs as well. © 2015 IEEE.
About the journal
JournalData powered by TypesetCanadian Conference on Electrical and Computer Engineering
PublisherData powered by TypesetInstitute of Electrical and Electronics Engineers Inc.
ISSN08407789
Open AccessNo
Concepts (16)
  •  related image
    Algorithms
  •  related image
    Artificial intelligence
  •  related image
    Digital arithmetic
  •  related image
    Learning systems
  •  related image
    Optimal systems
  •  related image
    Real time control
  •  related image
    Speech recognition
  •  related image
    ASSEMBLY CODE
  •  related image
    Efficient implementation
  •  related image
    Fixed points
  •  related image
    FIXED-POINT DSP
  •  related image
    FLOATING POINTS
  •  related image
    MACHINE LEARNING APPLICATIONS
  •  related image
    Real-time implementations
  •  related image
    SOFTWARE PIPELINING
  •  related image
    Learning algorithms