18
250-0800-06, Version A
D I T E C H N E T W O R K S
w w w . d i t e c h n e t w o r k s . c o m
Introduction
Packet Voice Processor Commands Manual
1.4.2.4 Automatic Level Control
When voice calls are connected between two different VoIP endpoint devices, the
volume levels are often unbalanced with one side of the call too high or low for
comfortable listening. Ditech’s Automatic Level Control (ALC) technology
dynamically detects level imbalances and automatically adds up to 15 dB or
attenuates as much as needed to bring both sides of the call to a specified volume
level.
1.4.2.5 Enhanced Voice Intelligibility
Encoding voice using compressed codecs reduces the sharpness of speech that can
lead to poor voice intelligibility. With an increasing number of IP borders and multiple
transcoding points across a network, providers must compensate for this loss in
fidelity. Ditech’s Enhanced Voice Intelligibility (EVI) feature improves the quality of
speech by boosting the critical speech formants within the audio to allow users to
more easily understand the voice within the call. The EVI algorithm improves the
intelligibility of speech without distorting or amplifying the signal.
Note
VQA technology is further described in detail in
1.4.3 Intelligent Packet Restoration (IPR)
IPR is an advanced feature that reconstructs missing packets within a VoIP packet
stream. With the capability to analyze and process the received voice packets, the
IPR feature uses a predictive speech model to reconstruct a missing packet’s voice
payload and rebuild it upon packet play out. This offers an enhanced capability to
support high quality voice over IP, even in congested IP networks with substantial
packet delay and loss. IPR is supported with the codecs for G.711, G.726, G.729,
and iLBC.
Intelligent Packet Restoration (restoration made through substitution of a
interpolated or modeled packet)
Jitter Buffer to correct for jitter that may be caused by network congestion,
timing drift, or route changes
With compressed media streams, transcoding and voice quality processing are
simultaneously performed. The following are the modes of operation:
Decode-Enhance-Encode to the same format.
Decode-Enhance-Encode to a different format. This mode combines
transcoding and voice quality processing. If the source or target format is
G.711 or G.729, then the decode or encode step may be omitted.
Note
IPR technology is further described in detail in
.
