Multimode Speech Coding - Digital Speech: Coding for Low Bit Rate Communication Systems - page 310

1

voiced

0.5

0

unvoiced

− 0.5

s(n)

unvoiced

s(n)

− 1

0 1000 2000 3000 4000 5000 0 1000 2000 3000 4000 5000

samples samples

(a) (b)

Figure 9.20 (a) Zero-crossing rate and (b) Voicing decision of the initial classification

1

200

low-band to high-band energy ratio

tracked energy

voiced

0.5

100

zero-crossing rate

0

0

unvoiced

− 0.5

− 100

s(n)

s(n)

1

−

− 200

0 1000 2000 3000 4000

0 1000 2000 3000 4000

samples

samples

Figure 9.21 Voicing metrics of the initial classification

signal is shifted down for clarity. Figure 9.20b depicts the voicing decision
made by the initial classification. Figure 9.21 depicts the three metrics used
and the final voicing decision over the same speech segment.
Even though the plosives have a significant amount of energy at high
frequencies and a high zero-crossing rate, synthesizing the high energy
spikes of the plosives using ACELP instead of noise excitation improves
speech quality. Therefore we need to detect the plosives, which are classified
as unvoiced by the initial classification, and switch them to ACELP mode.
A typical plosive is depicted at the beginning of the speech segment in
Figure 9.20b.