Acoustic signal processing


SPLab - DSP Laboratory

Subjects studied here: Digital filters, Digital signal processing, Digital signal processors, Speech processing
Person responsible: Ing. Petr Sysel, Ph.D.
Area of study: Human-machine interaction, Non-orthogonal signal representation, Acoustic signal processing

The laboratory equipment was acquired partly from the resources of the faculty and partly from FRVŠ Project 2890/F1a/2010 focused on innovation of the laboratory course Signal Processors. The laboratory has 25 workstations with modern computer technology, projection technology and 5-channel sound distribution. Each workstation is equipped with a MobilePREexternal sound card  with low latency from M Audio, an ATR25 stereo condenser microphone from Audio-Technica and K66 headphones from AKG. The laboratory is also equipped with DSO1022A oscilloscopes and 33220A functional generators from Agilent, DSP56858EVM development kits with signal processors from Freescale, TMDSDSK6416 and TMDSDSK6713 development kits with signal processors from Texas Instruments, and other laboratory instruments.

Thanks to the specified equipment, the laboratory provides background for the whole range of ​​digital signal processing and training courses focused largely on digital signal processing take place there. It is possible to design one- or two-dimensional digital filters and make their simulations in the Matlab computing environment. The Matlab environment is also used for the implementation of speech processing methods, such as voice activity detection, speech synthesis from text (Text-To-Speech), speaker recognition, speech recognition, etc. Debugged algorithms are implemented in signal processors with Harvard architecture or a very long instruction word (VLIW) architecture and tested on real signals in real time.

The research projects realized in the laboratory comply with the specialization of the laboratory.

  • GA102/07/1303 Nonlinear Methods for Speech Enhancement, 2007 – 2009.
  • FT-TA2/072 Research and Application of Methods of Time-Frequency Analysis for Speech Therapy, 2005 – 2009.
  • FD-K125 Application of Digital Speech Separation in Communications Technologies, 2001 – 2003.
  • FT-TA3/011 Research and Development of an Internet Telephone Exchange, 2006 – 2009.

After teaching hours, the laboratory equipment is available to students for dealing with individual projects, bachelor’s or master’s theses. Students can test and verify in practice their theoretical knowledge gained in individual subjects.

Specific devices:

In addition to the more or less standard laboratory equipment, the laboratory provides development kits for embedded systems and systems on module (SoM):

    • Development kit for PXA270 processor with ARM architecture from Marvell.
    • SoM – CPU 512 MHz, FLASH 64 MB, SDRAM 256 MB, audio, Ethernet 10/100 Mbps.
    • MMC/SD socket, CF socket, IDE PATA, PCMCIA, 2x USB, 1x OTG, 2x PS2, VGA, LCD 5,7‘‘ 640×480.
    • GNU/Linux 2.6, Opie, WinCE 6.0.
    • Processor module (COM – Computer On Module) used as a control unit in an Internet telephone exchange within the project FT-TA3/011.
  • OMAP35x Evaluation Module
    • Development kit for OMAP3530 dual-core processor with ARM architecture from Texas Instruments.
    • CPU – ARM Cortex-A8 720 MHz / DSP C64x+ 520 MHz; LPDDR 256 MB / FLASH 256 MB.
    • Audio, MMC/SD socket, USB 2.0, Ethernet 10/100 Mbps, DVI, S-Video In/Out.
  • ColdFile M547xEVB
    • Development kit for MCF547x and MCF548x processors with ColdFire architecture from Freescale.
    • SoM – CPU MCF5485 266 MHz. SDRAM 64 MB, 16 MB FLASH.
    • 4x USB, 1x OTG, CF socket, 2x PCI slot, 2x Ethernet 10/100 Mbps, VGA.

The laboratory also has a 16821A logic analyzer with a pattern generator from Agilent, used for development or testing of devices:

  • Logic analyzer: 34 channels, 16 MB, 500 MHz / 1.0 GHz.
  • Generator: 48 channels, 500 MHz.

« back to list