Telos Zephyr Xstream - Manual (Page 139)

USER’S MANUAL
Section 6: AUDIO CODING REFERENCE 127
AAC AAC-LD Layer-3 Layer-2 G.722
Delay :
Stereo (128kbps):
Sample Rate: --
32 kHz 242 msec 96 msec 452 msec 332 msec --
48 kHz 190 msec 68 msec 326 msec 220 msec --
Mono (64kbps):
Sample Rate:
16 kHz -- -- -- 590 14 msec
24kHz --
-- -- 402 msec --
32 kHz 284 msec 88 msec 390 msec 308 msec --
48 kHz 240 msec 56 msec 316 msec 208 msec --
Stereo (64kbps):
AAC @ 32kHz SR 404 msec -- -- -- --
Coding "Power"
16:1 - 18:1 12:1 12:1 6-8:1, 12:1 in
Joint stereo 4:1
Number of Bands
2048 2048 576 32 2
Notes:
* Layer-2 mono or discrete stereo @ 48 kHz sample rate yields 7.8 kHz audio response at 56 kbps and 9.8 kHz at 64 kbps
Delay will vary depending on ISDN network delay and other factors. Typical figures for encode-to-decode delay are given (network delay is not
included).
Frequency response is for swept sine test; response with program material will vary owing to the dynamic nature of the coding process.
As you can see from above, the Zephyr Xstream offers such a wide range of options for several
reasons:
• To optimize the tradeoff between delay and quality (see below);
• To offer compatibility with the large variety of equipment deployed in
existing installations.
• To offer a range of delay vs. quality to accommodate different
programming scenarios.
Delay vs. Quality
Looking at the chart, one thing that should be apparent is that there generally is a trade‐off
between delay and audio performance. Layer 3’s excellent audio performance requires a
significant delay. This is because some of its power comes from the ability to analyze the audio
over a relatively long period. Layer II requires the next longest delay, followed by AAC, AAC‐LD
and G.722. AAC‐LD and G.722 have minimal delay.