Glossary

Acoustic echo is caused by direct and indirect feedback from the speaker(s) to the microphone(s). An Acoustic Echo Canceller electronically removes both the direct coupling and the reflected echo, enabling true full-duplex hands-free telephony for both mobile phones and desktop speakerphones. There are many AEC algorithms to choose from and ...

The devices we use today for communications generate noise - it can be the A/C supply connected to them, the fans they have internally to keep their CPUs cool, or a myriad of other such "noise generators". While this noise may not be loud, the microphones will usually be close ...

  CBR is a technique for encoding voice and video which forces the output bitstream to stay within a specific bitrate constraint in a short timeframe, which is usually set to a single second. This reduces the flexibility of the encoding process to invest more bits in complex segments. CBR ...

When the conference room is not acoustically designed, reverberation in the room will be picked by the microphone and reflected to the loud speaker at the other side, which can lead to a very low quality and “box”-like voice. The Room De-Reverberation algorithm reduces the magnitude of the different audio ...

Forward Error Correction (FEC) is a type of correction code, which allows the receiver to compensate for loss packets without requiring any information to be sent in real-time by the sender.According to the FEC mechanism, the sender of the media stream sends redundant data as part of the stream. This ...

In a conference room there are situations in which the person speaking is not close enough to the microphone. For these cases you will usually place more than a single microphone within your microphone pod. This is done to improve the audio quality of the speaker in the room by ...

NetSense is RADVISION's bandwidth estimation and adaptation technology, enabling a high quality of experience even on unmanaged networks. NetSense estimates the available bandwidth in a network path, based on the delay in the video stream, therefore allowing to detect congestion in the network before it escalates to packet loss and ...

  VBR is a technique of encoding voice and video which allows the output bitstream to vary in the amount of bits per a given amount of time. This allows higher bitrates to be allocated to segments in the bitstream that are more complex than others. VBR is usually used ...

Today, everyone speaks about HD Voice - the ability to support voice signals that are over 8 KHz. While there are systems that actually support it today, there are ways to improve the audio quality of legacy systems that don't support HD Voice at all. This is done by extending ...