Customize “Tracking Area” for more accurate audio pick-up.

Users can now pre-set RM-CG’s beam-tracking zones using a color-coded system in the Web UI.

– Green represents the overall pick-up coverage of the microphone.

– Blue delineates “focus areas” for more localized sound pick up (up to 2 areas)

– Red creates an “exclusion area”, cancelling audio pick-up where specified (up to 2 areas)

Beamforming is a technology by which a directivity response pattern is formed using multiple microphones. The RM-CG can form up to four beams simultaneously. Narrowing the directivity of each beam and centering the beam on each speaker’s mouth enables this system to pick up sound while avoiding the negative impact of reverberation or ambient noise. Moreover, tracking with four beams offers the convenience of not having to set the sound collection range, and it prevents clipping of sound when a person begins to speak.

The Echo Canceller estimates the overall sound (including reflected sound) transmitted between speaker and microphone and reduces it (including reflected sound), thus preventing the voice from echoing back to the speaker. When the overall sound transmission is estimated perfectly, echo will not be generated. However, if the transmitted sound changes rapidly, the system will not adapt quickly enough and the likelihood of echo will increase. As mentioned above, the directions of the Multi-beam Tracking are constantly changing, but Yamaha’s Echo Canceller uses an algorithm capable of responding quickly to any changes. As a result, high echo cancellation performance is achieved despite the use of Multi-beam Tracking.

Noise Reduction used in teleconferencing systems estimates the noise contained in a signal from the signal itself and attenuates its components. It enables clear sound collection by detecting steady noise and attenuating only those components.

The RM-CG analyzes the components of picked-up sound in real time, then estimates and removes any unpleasant reverberation components in the sound based on an acoustic model, enhancing the clarity of the picked-up sound.

The RM-CG reduces the fluctuation in voice levels sent to other parties by automatically adjusting the gain depending on the levels of various voices picked up. In addition, HVAD (Human Voice Activity Detection) works to adjust the voice level only. In this way, HVAD prevents unpleasant sounds from being generated without unnecessarily raising the level of unwanted sound.

General gain-sharing automatic mixing is often used when a microphone is mounted close to the speaker. When applied to the RM-CG, it tends to mix the four beams in a similar way. For this reason, the RM-CG uses automatic mixing optimized for the installation conditions, which demonstrates strong performance even in environments such as ceiling installations or rooms with a lot of reverberation.