6/23/09 J-DSP: An Online DSP Laboratory Overview J-DSP J-DSP - - PDF document

6 23 09
SMART_READER_LITE
LIVE PREVIEW

6/23/09 J-DSP: An Online DSP Laboratory Overview J-DSP J-DSP - - PDF document

6/23/09 J-DSP: An Online DSP Laboratory Overview J-DSP J-DSP Editor Editor J-DSP blocks to generate, process and understand analog signals. Application of J-DSP to courses in Arts, Media and Engineering Department at Arizona State


slide-1
SLIDE 1

6/23/09 1

J-DSP Editor

Sponsored by NSF Awards 0817596, NSF-DUE-CCLI-080975 NSF Program CCLI Phase 3 Award Started Apr. 2008 – Apr. 2013 involves 8 universities Also core software used in an NSF CRCD 2004-2006

Mohit Shah 1

FIE WORK IN PROGRESS 2005

J-DSP: An Online DSP Laboratory

J-DSP Editor

 J-DSP blocks to generate, process and understand

analog signals.

 Application of J-DSP to courses in Arts, Media and

Engineering Department at Arizona State University.

Mohit Shah 2

FIE WORK IN PROGRESS 2005

Overview

J-DSP Editor

 Parameters

  • Signal Type – Sine, Square, Triangular, Exp.
  • Amplitude
  • Analog Frequency
  • Phase

 Output: A string symbolic representation of the chosen signal.

  • For e.g. - 4sin(50t + 60) or 16exp(-20t)

Mohit Shah 3

FIE WORK IN PROGRESS 2005

Analog Signal Generator

J-DSP Editor

 Linear combination of signals

Mohit Shah 4

FIE WORK IN PROGRESS 2005

Analog Signal Processing

slide-2
SLIDE 2

6/23/09 2

J-DSP Editor

 Sampling and Quantization

Mohit Shah 5

FIE WORK IN PROGRESS 2005

Analog Signal Processing (Contd.)

J-DSP Editor

 Extend the algorithm to allow the user to enter any

expression mathematically, automatically parse the expression and generate the corresponding signal.

 Expand the current library of blocks to include more

blocks that help understand the behavior of analog signals.

Mohit Shah 6

FIE WORK IN PROGRESS 2005

Future Work

J-DSP Editor

 Sound Sensing and Analysis

  • Algorithms to extract important features from audio,

speech signals or environmental sounds based on their content.

  • Some examples of common features -
  • Pitch
  • Temporal Energy
  • Harmonicity
  • Spectral Centroid
  • Mel Frequency Cepstral Coefficients.
  • Tonality Index

Mohit Shah 7

FIE WORK IN PROGRESS 2005

Arts, Media and Engineering

J-DSP Editor

 Track the fundamental frequency or pitch of a given

frame.

Mohit Shah 8

FIE WORK IN PROGRESS 2005

Pitch Estimation

slide-3
SLIDE 3

6/23/09 3

J-DSP Editor

 Probabilistically measure whether the given frame

exhibits a harmonic frequency structure or not.

 Harmonicity averaged over a number of frames is an

important feature to study environmental sounds.

Mohit Shah 9

FIE WORK IN PROGRESS 2005

Harmonicity

J-DSP Editor

 Indicates the location of the ‘center of mass’ of the

spectrum.

 Perceptually gives an impression regarding the

‘brightness’ a given sound.

Mohit Shah 10

FIE WORK IN PROGRESS 2005

Spectral Centroid

J-DSP Editor

 Used in psychoacoustic models to determine whether a

given sound exhibits a tone-like or noise-like characteristic.

Mohit Shah 11

FIE WORK IN PROGRESS 2005

Tonality Index

J-DSP Editor

 Calculate the average energy of a given frame. Use

threshold to mark the frame as voiced or unvoiced.

Mohit Shah 12

FIE WORK IN PROGRESS 2005

Temporal Energy

slide-4
SLIDE 4

6/23/09 4

J-DSP Editor

 Integrate J-DSP with Processing to create interactive

simulations and programs to aid interdisciplinary students in AME.

Mohit Shah 13

FIE WORK IN PROGRESS 2005

Future Work

J-DSP Editor

 Professor Andreas Spanias, ASU.  Professor Michael Stiber, UWB.  Assistant Professor Harvey Thornburg, AME, ASU.  National Science Foundation.

Mohit Shah 14

FIE WORK IN PROGRESS 2005

Acknowledgements

Thank You