Computer Sound Synthesis 2 - PowerPoint PPT Presentation

1 / 19
About This Presentation
Title:

Computer Sound Synthesis 2

Description:

Title: PowerPoint Presentation Author: John Woodruff Last modified by: Gary Kendall Created Date: 1/3/2006 3:39:57 PM Document presentation format – PowerPoint PPT presentation

Number of Views:28
Avg rating:3.0/5.0
Slides: 20
Provided by: JohnW369
Category:

less

Transcript and Presenter's Notes

Title: Computer Sound Synthesis 2


1
Computer Sound Synthesis 2
  • MUS_TECH 335 Selected Topics

2
Filters continued
3
Complex NumbersComplex Plane
imaginary
x yi
y
real
x
Cartesian coordinates
4
Im
polar form A q
A
q
Re
A x2 y2
y x
q arctan( )
5
Im
A
y A sin q
q
Re
x A cos q
Relationship of polar form to Cartesian form.
6
Complex Numbers
Addition (abi) (cdi) (ac) (bd)i
Multiplication A1?q1 A2?q2 A1A2?(q1
q2) (abi)(cdi) ac adi bci bdi2 (ac -
bd) (ad bc)i
.
-1
7
Meaning of i?
  • i represents a 90 degree rotational shift
  • i i i2 -1 represents a 180 degree
  • rotational shift

.
8
imaginary
z-plane
unit circle
frequency
real
Nyquist
0
-1/2 SR
-3/2 SR
3/2 SR
1/2 SR
9
Geometric Interpretation of Magnitude Response
img
z-plane
pole
P1
zero
Z1
real
Z1 Z2 P1 P2
H proportional to
10
First-order Non-recursive Filter
zero
a
high-pass
a 1
Z
f
0
SR/2
z-plane
a
a -1
low-pass
Z
f
0
SR/2
z-plane
11
First-order Recursive Filter
pole
low-pass
a
P
b .9
f
0
SR/2
Z-plane
high-pass
a
P
b -.9
f
0
SR/2
Z-plane
12
Take a look at Movie Demonstrations
  • http//www.ece.msstate.edu/hagler/Aug1996/011/cd/
    Demos/Z2freq/index.htm

13
2nd-order Filters
2-zero filter
a1
k
a2
z-1
z-1
non-recursive
k
2-pole filter
z-1
z-1
b1
b2
recursive
14
2-zero band-reject
a
f
0
SR/2
2-pole band-pass
(xyi)
a
f
0
(x-yi)
SR/2
-1
(x yi)(x - yi) x2 - xyi xyi - y2i2 x2 y2
15
low Q
high Q
f
freq BW
Q
Q increases as pole approaches the unit circle

r
Used in SuperCollider

BW freq
1

RQ

Q
q
16
Visit Pole-Zero Filter Design Applet
  • http//www.earlevel.com/Digital20Audio/PoleZero.h
    tml

17
two-pole filter
y(n)
k
x(n)
z-1
z-1
-b1
-b2
difference equation
y(n) k x(n) - b1 y(n-1) - b2 y(n-2)
b1 2r cos q b2 -r2
The coefficients depend on the position of the
poles, here expressed in polar form, q and r.
18
Programming Implementation
output
input
old1
old2
z-1
z-1
k
feedback loop
-b1
-b2
Program Flow
output k input - b1 old1 - b2 old2 old2
old1 old1 output



19
Biquad Filter
1
a1
a2
out
k
z-1
z-1
b1
b2
combined IIR/FIR 2nd-order
Write a Comment
User Comments (0)
About PowerShow.com