AudioCodes Mediant 3000, User Manual

Page: 189 / 872

Version 6.6 189 Mediant 3000
User's Manual 13. Media
Approximate AMD Detection High Sensitivity (Based on North American English)
AMD Detection
Sensitivity
Performance
Success Rate for Live Calls Success Rate for Answering Machine
0 (Best for
Answering
Machine)
72% 97%
1
77% 96%
2 79% 95%
3 80% 95%
4 84% 94%
5 86% 93%
6 87% 92%
7 88% 91%
8 (default) 90% 89%
9
90% 88%
10 91% 87%
11 94% 78%
12 94% 73%
13 95% 65%
14 96% 62%
15 (Best for Live
Calls)
97% 46%
Note: The device's AMD feature is based on voice detection for North American
English. If you want to implement AMD in a different language or region, you
must provide AudioCodes with a database of recorded voices in the language
on which the device's AMD mechanism can base its voice detector algorithms
for detecting these voices. The data needed for an accurate calibration
should be recorded under the following guidelines:
•
Statistical accuracy: The number of recordings should be large (i.e.,
about 100) and varied. The calls must be made to different people, at
different times. The calls must be made in the specific location in which
the device's AMD mechanism is to operate.
•
Real-life recording: The recordings should simulate real-life answering of
a person picking up the phone without the caller speaking (until the AMD
decision).
•
Normal environment interferences: The environment should almost
simulate real-life scenarios, i.e., not sterile but not too noisy either.
Interferences, for example, could include background noises of other
people talking, spikes, and car noises.