Telos Zephyr Xstream User Manual page 132

•
No dynamic block switching is used because the required look‐ahead delay
is too big. The temporal problem that causes pre‐echoes is handled by the
Temporal Noise Shaping module.
•
The "window shape" of the spectral filter is enhanced to be adaptive.
Normally, the shape is a simple sine curve, but AAC‐LD can use a shape that
has a lower overlap between the bands. This significantly improves
performance with transients, without adding any delay.
Amazing Performance
Low delay would not be useful if the quality was not acceptable, of course. So how does AAC‐LD
stack‐up? The results are shown below: AAC‐LD is clearly better than Layer‐3 for half of the test
items, and as good for the remaining half. Not bad considering that AAC‐LD has a fraction of the
delay.
Performance comparison of Layer 3 and AAC-LD. A = AAC-LD, B = Layer 3.
Results here are very good for AAC‐LD. Test items es01‐03 are speech. Since most mono
remote broadcasts are speech, you can expect audio quality even better than with the familiar –
and already perfectly useful – Layer‐3. Pop music is the test item sc03 and performance is good
here as well. The one slight trouble spot is the glockenspiel at sm02...
We expect you will find this mode very useful.
ISO/MPEG LAYER-3
After extensive testing by broadcasting organizations around the world under the direction of
the CCIR, it had been designated as the most powerful of the three audio coding systems
USER'S MANUAL
Section 6: AUDIO CODING REFERENCE 120