| Version 1.2 |
Requires MATLAB
6.0 (R12) or later |
|
|
|
Linearly
maps all elements of the input matrix on to the output
range -1:+1. Useful for scaling audio WAV data which
has been read with wavin
using the 'raw' scale option. |
|
| File
format: |
m-file |
|
Editable
source code:
|
yes
|
| Utilises
non-editable functions: |
no |
|
Platform:
|
PC/Windows
|
| Required
MATLAB Toolboxes: |
none
(except core MATLAB) |
| Demo
version limitations: |
p-code
only (non-editable); 2000 sample length limit (then
silence) |
|
|
|
|
|
|
|
| Input: |
| x |
matrix of input data. For example, if x is read from
a WAV file using wavin,
then the dimensions will be (Number of samples * Number
of Channels).
|
|
| Output: |
| y |
matrix
of output data, linearly normalised to the amplitude
range -1:+1.Same dimensions as input matrix.
|
|
|
|
|
The
normalization is applied across the entire data set
(i.e. across all channels), using a range determined
by the minimum value across the entire data set and
the maximum value across the entire data set.
|
|
|
|
|
|
|
|
| The
following example demonstrates how to read a WAV file
using wavin
with the 'raw'
scale setting, then normalize it using wavscale: |
| |
| Read
a 4-channel WAV file named 4channel.wav
located in
the directory ..\WAVFiles using the 'raw'
scale setting and explore the data range: |
| |
[x,fs]=wavin('..\WAVfiles\4channel.wav'); |
| |
>>
[min(x) max(x)] |
| |
ans
= -10254 -12067 -18884 -31783 |
| |
22258 22219
26018 19188 |
| |
|
| Now
normalize using wavscale and explore the resulting data
range (observing that the results are confined to -1:+1
across all channels): |
| |
y=wavscale(x); |
| |
>>
[min(y) max(y)] |
| |
ans
= -0.2551 -0.3178 -0.5537 -1.0000 |
| |
0.8699 0.8685
1.0000 0.7637 |
| |
|
