Handbook for Sound Engineers

154 Chapter 7

7.2.2.4 Center Time (ts) (Kürer)

For music and speech performances, the center time ts is
a reference value for spatial impression and clarity and
results at a measuring position from the ratio between
the summed-up products of the energy components of
the arriving sound reflections and the corresponding
delay times and the total energy component. It corre-
sponds to the instant of the first moment in the squared
impulse response and is thus determined according to
the following ratio:

(7-22)

The higher the center time ts is, the more spatial the
acoustic impression is at the listener’s position. The
maximum achievable center time ts is based on the
optimum reverberation time. According to Hoffmeier,^12
there is a good correlation between center time and
intelligibility of speech with a frequency evaluation of
four octaves between 500 Hz, 1000 Hz, 2000 Hz, and
4000 Hz.
For music, the desirable center time ts is ts|70 to
150 ms with a 1000 Hz octave, and for speech ts|60 to
80 ms with four octaves between 500 and 4000 Hz.

7.2.2.5 Echo Criterion (EK) (Dietsch)

If we look at the build-up function of the center time
ts(W):

(7-23)

where,
the incoming sound reflection n = 0.67 with speech and
n = 1 with music.

Comparing it with the difference quotient

(7-24)

we can discern echo distortions for music or speech
when applying values of = 14 ms for music and
= 9 ms for speech, ascertained by subjective

tests.^13 The echo criterion depends on the motif. With

fast and accentuated speech or music, the limit values

are lower.

For 50% (EK50%) and 10% (EK10%), respectively, of

the listeners perceiving this echo, the limit values of the

echo criterion amount to:

• Echo perceptible with music for EK50%t1.8;
  EK10%> 1.5 for two octave bands 1 kHz and 2 kHz
  mid frequencies.


• Echo perceptible with speech for EK50%t1.0;
  EK10%> 0.9 for one octave band 1 kHz.

7.2.2.6 Definition Measure C 50 for Speech (Ahnert)

The definition measure C 50 describes the intelligibility

of speech and also of singing. It is generally calculated

in a bandwidth of four octaves between 500 Hz and

4000 Hz from the tenfold logarithm of the ratio between

the sound energy arriving at a reception measuring posi-

tion up to a delay time of 50 ms after the arrival of the

direct sound and the following energy:

(7-25)

A good intelligibility of speech is generally given when

C 50 t0dB.

The frequency-dependent definition measure C 50

should increase by approximately 5 dB with octave

center frequencies over 1000 Hz (starting with the

octave center frequencies 2000 Hz, 4000 Hz, and

8000 Hz), and decrease by this value with octave center

frequencies below 1000 Hz (octave center frequencies

500 Hz, 250 Hz, and 125 Hz).

According to Höhne and Schroth,^14 the limits of the

perception of the difference of the definition measure

are at 'C 50 |±2.5 dB.

An equivalent, albeit less used criterion, is the

degree of definition D, also called D 50 , that results from

the ratio between the sound energy arriving at the recep-

tion measuring position up to a delay time of 50 ms

after the arrival of the direct sound and the entire energy

(given in %) is

(7-26)

The correlation with the definition measure C 50 is deter-

mined by the equation

ts

tiEi

i

¦

Eges

----------------=

ts W

pt n t td

0

W

³

pt

n

td

0

W

³

=------------------------------

EK W

'ts W

'tE

---------------=

'tE

'tE

C 50 10

E 50

Ef–E 50

= log©¹§·--------------------- dB

D

E 50

Ef

--------=