Neural networks - Signals, Processes, and Systems - page 452

12, 00

“neuronal networks” astotal acoustic
signal in the time range

6, 50

1, 00

-4, 50

- 10, 00

50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850

Y/t-Grafi k 0

ms

0, 8
0, 7
0, 6
0, 5
0, 4
0, 3
0, 2
0, 1
0, 0

“ neuronal networks” astotal acoustic
signal in the frequency range

Near- periodic signals like vocals
and tones create a line - shaped
spectrum

0 250 500 750 1000 1250 1500

Y /t-Grafi k 0

Hz

“neuronal networks” as
superposition of acoustic signal
parts in the time range (excerpt )

10,00

Frequency - time- landscape of
“neuronal networks ”
The near - periodic parts of the
vocals create a “bed of nails” -
pattern. In this form, the acoustic
pattern is fed to the brain and
compared to the other patterns

7,50

5,00

2,50

0,00

0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 3250 3500 3750

Hz

Illustration 292: About the physical basics and problems of speech recognition
Via microphone, the term “neuronal networks” was filed as acoustic signal. Above, this signal is displayed
in its entirety in the time- and frequency range. Because of the mentioned reasons, for the purpose of
acoustic signal recognition, the total signal has to be dissected in time- or frequency parts (see also
chapter 4).

Due to the near- periodic parts in voice (vocals) or music (tones), a pattern resembling a “bed of nails” all
in all results in the frequency range. The identification respectively association of this acoustic signal in
the brain can be approximately explained as pattern recognition of such “beds of nails”. This especially is
valid for electronic speech recognition by the means of computers.