Talk:Digital Signal Processing Software: Difference between revisions

From OLPC
Jump to navigation Jump to search
 
(4 intermediate revisions by 2 users not shown)
Line 11: Line 11:




# Explore OpenSound Toolkit - explore how sound objects work, what filters it already provides, how to configure new filters and how viable (in terms of extra packages required) it would be to use OpenSound toolkit on the XO.
# <s>Explore OpenSound Toolkit - explore how sound objects work, what filters it already provides, how to configure new filters and how viable (in terms of extra packages required) it would be to use OpenSound toolkit on the XO.</s> Not viable for XO
# Get sugar-jhbuild running - [[Sugar_with_sugar-jhbuild]]
# <s>Get sugar-jhbuild running - [[Sugar_with_sugar-jhbuild]]</s> Done
# Explore TamTam [[TamTam:Source_Code_Overview]]
# Explore TamTam [[TamTam:Source_Code_Overview]]
# With the help of 1 and 3 define goals of application and general UI layout
# With the help of 1 and 3 define goals of application and general UI layout
# Read up on how to do basic sound processing in python
# Read up on how to do basic sound processing in python
# Use MATLAB to test out mathematical operations.
# <s>Use MATLAB to test out mathematical operations.</s>




Line 22: Line 22:
* Implementation basics
* Implementation basics


# <s>gstreamer</s> CSound server model will be used for all signal processing. Check TamTam to see how CSound is used within the application.
# gstreamer
# cairo
# cairo
# pygtk
# pygtk




* Attended by Hemant, Asim and Arjun
* Attended by Hemant, Assim and Arjun

== Update of meeting held on 23Sep07 ==

# Tested filters like Multi Echo, Delay.
# Experimented with Sampling rate, Frequency spectrum of sound files to develop a basic intution about DSP using MATLAB.
# To fix all sound effects that will be provided in the prototype and find the mathematical equation for the filters.


* Attended by Hemant, and Assim

== Update of meeting held on 25Sep07 1330Hrs IST ==

# Getting familiar with basic CSound structure.
# Played around with few CSound programs to have better understanding.


* Attended by Hemant, and Assim

== Update of meeting held on 27Sep07 1125Hrs IST ==

# Filters to be implemented in first developmental phase
## Echo
### Number of echoes
### Delay
### Volume
### Feedback
## Flangers
### Overall Characteristics
#### Source percentage
#### Flanger Percentage
#### Feedback percentage
### Flanger Characteristics
#### Variable Delay
#### Frequence
#### Fixed Delay
#### Sine Modulator
#### Triangle Modulator
## Pitch
### Scale Percentage/Semitone and FineTone
### Preserve Tempo
### Overlap
### FFT Size
## Reverse
## Mechanize
### Frequency
### Modulator
#### Sine
#### Triangle
#### Square

Next Steps:
Explore CSound Filter libraries and algorithms for above filters

*Attended by Assim and Hemant

Latest revision as of 05:59, 27 September 2007

this is a Geode

The hardware is slow. You need to take advantage of what little you have, rather than wasting CPU cycles (including memory access -- there is barely any cache) on thick abstraction layers and other fluff. You get MMX, 3dNow!, a bit of SSE, a bit of SSE2, and some Geode-specific instructions.

Don't be afraid of assembly. Assembly is good for you. It builds character.

AlbertCahalan 02:37, 23 September 2007 (EDT)

Meeting minutes 10Sep07 , 1330hrs

  • Things TODO:


  1. Explore OpenSound Toolkit - explore how sound objects work, what filters it already provides, how to configure new filters and how viable (in terms of extra packages required) it would be to use OpenSound toolkit on the XO. Not viable for XO
  2. Get sugar-jhbuild running - Sugar_with_sugar-jhbuild Done
  3. Explore TamTam TamTam:Source_Code_Overview
  4. With the help of 1 and 3 define goals of application and general UI layout
  5. Read up on how to do basic sound processing in python
  6. Use MATLAB to test out mathematical operations.


  • Implementation basics
  1. gstreamer CSound server model will be used for all signal processing. Check TamTam to see how CSound is used within the application.
  2. cairo
  3. pygtk


  • Attended by Hemant, Assim and Arjun

Update of meeting held on 23Sep07

  1. Tested filters like Multi Echo, Delay.
  2. Experimented with Sampling rate, Frequency spectrum of sound files to develop a basic intution about DSP using MATLAB.
  3. To fix all sound effects that will be provided in the prototype and find the mathematical equation for the filters.


  • Attended by Hemant, and Assim

Update of meeting held on 25Sep07 1330Hrs IST

  1. Getting familiar with basic CSound structure.
  2. Played around with few CSound programs to have better understanding.


  • Attended by Hemant, and Assim

Update of meeting held on 27Sep07 1125Hrs IST

  1. Filters to be implemented in first developmental phase
    1. Echo
      1. Number of echoes
      2. Delay
      3. Volume
      4. Feedback
    2. Flangers
      1. Overall Characteristics
        1. Source percentage
        2. Flanger Percentage
        3. Feedback percentage
      2. Flanger Characteristics
        1. Variable Delay
        2. Frequence
        3. Fixed Delay
        4. Sine Modulator
        5. Triangle Modulator
    3. Pitch
      1. Scale Percentage/Semitone and FineTone
      2. Preserve Tempo
      3. Overlap
      4. FFT Size
    4. Reverse
    5. Mechanize
      1. Frequency
      2. Modulator
        1. Sine
        2. Triangle
        3. Square

Next Steps: Explore CSound Filter libraries and algorithms for above filters

  • Attended by Assim and Hemant