Conference Audio Systems Design

Mar 1, 2001 12:00 PM, Michael Pocino

HIGH-QUALITY AUDIO IS VITAL IN CONferencing and distance learning systems. Poor audio can make a conference fatiguing and even unintelligible. While there are many things that contribute to a good conferencing system, nothing works better than starting with good acoustics.

A room with lots of hard surfaces such as walls, floors, ceilings, windows and tables can be too reverberant. These hard surfaces are often parallel to each other, causing sound to bounce around the room before decaying. A microphone in this type of sound environment picks up everything and sends it out to speakers or to other meeting places your system is connected with. The original sound gets the same treatment as a reflection of the sound because the microphone cannot differentiate between the two. All sounds reaching the connected conference room are from a single location (the loudspeaker on their end). The result is that the conference room on the receiving end hears everything original audio and reflections at one level (the loudspeaker's volume).

You can see how this would ruin the meeting for anyone at a remote location. Since remote/teleconferencing is becoming more prevalent, you need to know how to make audio sources less noisy.

There are a number of ways to correct reverberant rooms. Your first recourse is acoustic treatment: acoustical ceiling tiles, carpeted floors, sound-absorbing panels on the walls and draperies over the windows. If you don't have the budget for these things, try tacking a few blankets to the walls and tossing some throw rugs on the floor. You'll be surprised how much quieter the room gets.

CORRECT MICROPHONE USAGE

Ideally, we want to get a 25 dB signal-to-noise ratio at the mic in a conference room system. A problem to overcome on any system's transmit path is the fact that noise is transmitted with speech, and adding gain won't help that situation. Keeping mics away from noise is the key. It's crucial to maintain an appropriate number of mics, enough that everyone can be heard adequately, without using too many, which can add background noise. A good rule of thumb is to use one mic for every two or three conference speakers.

Low-profile boundary mics that sit on a tabletop are attractive and work well for most corporate conference rooms. Push-to-talk mics can also be used, but be sure to specify models that do not mute the mic element. Push-to-talk mics should mute mic signals in the mixer, so they must be designed with separate conductors for the switch. Four such mics are beyerdynamic's MPC67RC, Crown's PCC170SW0, and Shure's MX392 and MX412D. Check with the manufacturer to see if they make any models with separate contact closures.

Wireless mics can also be used; however, movement challenges the operation of acoustic echo cancellers. Lavalier mics are better than handheld when it comes to wireless, and it is critical to keep them a fair distance from loudspeakers to avoid feedback problems.

Podium (or gooseneck) mics are also a good option in paper-laden conference rooms to avoid the sounds of rustling paper. The key here is to get the mics off the desktop, but still within the critical distance of 2 to 3 feet from the speaker. Microphones that are flush-mounted on the ceiling can also work but can create problems if they are too far away from the subject. They may also pick up vibrations from the HVAC system, loudspeakers or even people walking on the floor above.

If the boardroom has more than four microphones in use it is wise to use an automatic mic mixer. When a large number of mics are on at once, especially in a reverberant room, the room's gain into the audio system can become too high. This excessive gain results in feedback and can increase background noise in the audio system. The automatic mic mixer will ensure that a minimum number of mics are on at once.

CORRECT LOUDSPEAKER USAGE

Ideally, the loudspeaker should be 25 dB louder than the noise level of the room it is in. A common mistake in teleconferencing systems is the tendency to use only one or two loudspeakers to carry the audio from the other sites. When too few loudspeakers are used it is necessary to increase the volume, which places the room closer to potential feedback. Also, the listeners hearing the loudspeakers directly are in pain from the volume, while those who are farther away from them can barely hear what is going on.

Using an adequate number of loudspeakers will allow you to keep volume at a reasonable level throughout and will minimize the amount of loudspeaker audio that is picked up by mics. The best way to distribute loudspeakers is to place them in the ceiling (or mount them along the walls) and control them with a separate power amplifier. Also, try to isolate loudspeakers from mics as much as possible. Remember that a mic will pick up all sounds in its vicinity, and audio coming out of a speaker will be treated as just another voice in the room.

NOISE SUPPRESSION

In a multipoint conference system, the complexity is compounded as more noise sources are mixed together. It may be beneficial (or necessary) to reduce noise using electronic methods.

Noise suppression is an electronic process of removing noise from a corrupted signal. Generally, noise suppression consists of a method for distinguishing between the signal and the noise and a method for removing the noise. There are many different kinds of noise suppression, but they all have one basic goal: to make the output sound as much like the original, noise-free signal as possible.

The best way to reduce noise in a signal is to prevent it from entering the signal to begin with. This is accomplished by using good system design practices such as improving acoustics, using high-quality equipment and following good wiring practices. Despite all these efforts, there will always be some noise in the signal.

Note that noise suppression isn't necessarily a cure for noisy signals. A signal with 10 dB of noise suppression doesn't sound as good as a signal that has 10 dB less noise in the first place. But it will still sound much better than a signal with no noise suppression at all!

NOISE BEYOND YOUR CONTROL

Conference systems are often connected to a wide variety of remote locations, many of which may be noisy. Noise is frequently introduced by a poorly designed room on the other end of the conference. Even when all rooms are well-designed, noise may be introduced from sources beyond everyone's control. Common sources of noise include blowers and HVAC systems; road noise from cellular phones; in-band telephone hiss or hum that cannot be removed by the phone add; or noise introduced through the conference transmission system itself. Other sources are computers and projectors used in the conference room.

Since you don't have control of the other rooms or the communications network, your only choice is to remove the noise electronically. Ideally, we want to reach a signal-to-noise ratio of 25 dB. Other ways to increase SNR is to place mics closer to talkers, since the talker's SPL increases closer to the mic. Applying acoustical treatments to the room is also an option; however it might be expensive in a retrofit so it is usually best left for inclusion in the design phase.

The effects of noise are gradual, and listeners may get fatigued after listening to high levels for a long period of time. Even if the intelligibility isn't improved by noise suppression, the quality enhancement can make the conference much more pleasant and allow it to continue comfortably for a longer period of time.

MIXING CHANNELS TOGETHER

Even with moderate to low noise levels, the noise starts to add up when you mix signals together. This will limit the number of channels you can mix together, depending on the noise levels of each channel and your quality requirements. Noise suppression can effectively increase the number of channels you can mix together or simply improve the audio quality with the number of channels you have. If noise suppression is applied correctly, you can improve the signal quality and intelligibility just as much as if you had prevented the noise from entering the system acoustically. This process is best illustrated by the following example.

A conference room has eight non-gated microphones, which are on the table about 3 feet from the participants. The ambient noise level in the room is 40 dB SPL, and the level of speech picked up by the microphone is 73 dB SPL. The unprocessed signal-to-noise ratio for each channel is 33 dB.

When you mix the eight channels together without gating, the total noise level adds up to 49 dB (add 3 dB for every doubling of the number of channels). This means the total SNR is only 24 dB. If you apply noise suppression to the mix, the SNR is improved to 34 dB, but the signal doesn't sound as good as it would if the SNR was 34 dB without noise suppression. The mixed noise suppression is certainly an improvement, but you can do even better.

If you apply noise suppression to each individual channel before mixing, the algorithm has to remove less noise. Since you are applying noise suppression to a signal channel, you are starting with an SNR of 33 dB. After noise suppression, the SNR is improved to 43 dB. Now when you mix the channels together, the SNR is brought back to 34 dB. The key here is that the SNR has never been worse than 33 dB, so the mix sounds as good as a signal channel with a 33 dB SNR. That is, it is not a 23 dB SNR signal that has been improved with noise suppression to sound like 33 dB.

This works because removing noise from the channels individually prevents noise from one channel from getting mixed in with all the others. Since you are preventing noise from getting mixed in, the intelligibility doesn't suffer. For example, the noise in channels 2 through 8 is reduced before it gets mixed with channel 1. So the intelligibility of channel 1 is not harmed by the noise from the other channels. This is the same principle that makes an automatic mic mixer sound better than a single noise gate after a non-gated mixer.

MULTIPOINT CONFERENCING

The benefits of using noise suppression in multipoint conferencing systems are similar to the benefits you get when mixing channels together. Since the sites are mixed together, you can improve intelligibility and mix quality by reducing noise at each site. The benefits of noise suppression in this situation may be even more noticeable if one of the sites is very noisy.

Consider a simple system with two rooms (A and B), linked with a high quality audio conferencing channel. If a telephone caller on a noisy line is brought into the conference, the line noise corrupts the mix. Thus, the people in rooms A and B sound just as noisy as the telephone caller. By removing the noise before it gets mixed in, the people in room A and B sound much more intelligible, and the telephone signal has better perceived quality.

NOISE SUPPRESSION METHODS

Before we compare some of the noise suppression methods for audio and communications, please see the criteria in the sidebar, What Makes a Good Noise Suppression Algorithm?

If these criteria are met, the speech signal will not be distorted by the noise suppression process. Ideally, the algorithm will make no difference in audio quality except for the reduction of noise. You shouldn't know it was there until you turn it off and hear an increase in noise level. And though adequate noise reduction will occur, intelligibility won't suffer.

Noise Gates

Noise gates attenuate the telephone signal when there is no speech present and turn the gain back up when someone starts talking. Thus, they only remove noise during idle periods. At low noise levels, this is not noticeable since the noise is masked by speech. As the noise gets louder, its presence during speech becomes more noticeable and results in a loss of perceived quality. At moderate noise levels, there may be a noticeable whooshing sound as the gain is ramped up and down to allow the speech to pass through. This may result in a half-duplex feel because there is obviously some noise, but then the signal goes completely dead when there is no speech. If the noise levels get too high, the noise gate may have trouble deciding what is speech, and some speech may actually get cut off. Most noise gates have threshold adjustments which must be manually set for different noise levels for the best performance.

Speech Enhancement

Various methods of speech enhancement or emphasis work by trying to increase the perceived level of speech. They don't actually remove noise, but try to emphasize the speech parts of the signal. This is usually accomplished by enhancing the speech formats. If the level of the signal can be turned down (due to the increased speech level), noise is effectively removed. In general, these methods cannot improve the signal to noise ratio by very much (perhaps about 3 to 6 dB). Furthermore, intelligibility is not preserved very well because certain consonant sounds are not emphasized.

Spectral Subtraction

Spectral subtraction is a noise reduction method that makes an adaptive estimate of the noise spectrum and subtracts that from the signal. The noise estimate is subtracted all the time, both during speech and idle periods. However, there is an artifact associated with spectral subtraction called musical noise, which is a sort of trickling sound added while the noise is removed. It also adds a hollow, resonant quality to speech. This tends to be annoying and distracting.

In general, there is a tradeoff with spectral subtraction methods. You can get moderate noise cancellation with an unacceptable amount of musical noise or a good amount of noise cancellation with an unacceptable amount of speech attenuation. Spectral subtraction does a fair job of removing noise, but it adds many artifacts that may be more annoying than the noise itself.

Adaptive Digital Filtering

Adaptive digital filtering algorithms appear to be the answer. Figures 2a to 2d show remarkable gain reduction in acoustic echoes. Adaptive digital filters employ DSPs designed to differentiate between noise and program signal after converting the signal into the digital domain. Their effect is striking and is the current state of the art.

Acoustic Gain

In audio conferencing, the idea of acoustic gain is a little different than in sound reinforcement. In this application, we determine how much of the loudspeaker signal is being picked up by the microphone or how loud the loudspeaker signal is at the mic compared to local speech. In other words, acoustic gain is the difference between amplified and unamplified speech at the farthest listener. If we know the distances involved, it is a simple matter to calculate the needed acoustic gain, which is the minimum gain necessary for comfort and intelligibility, and the potential acoustic gain, which is the maximum level without feedback. The idea is to design a system so that PAG is greater than NAG (refer to Figure 3).

ACOUSTIC ECHO CANCELLATION

What happens when you have too much acoustic gain? If the acoustic gain of your system exceeds the acoustic echo canceller's capabilities, the acoustic echo cancellation will no longer adapt well to changing acoustic conditions in the room. This can result in increased noise suppression causing half-duplex communications, lack of convergence (residual echo heard all the time) and excessive feedback. As the acoustic gain increases, these problems will get worse and may make communications impossible.

Acoustic echo is most noticeable (and annoying) when delay is present in the transmission path. This happens primarily in long-distance circuits or in systems using speech compression (such as video conferencing or digital cellular phones). Even though the echo might not be as annoying when there is no delay (as with short links between conference rooms in the same building or distance learning over fiber-optic cable), it is still intrusive and can cause fatigue and listener stress.

Acoustic echo cancellers can be used in both narrow-band (3.5 kHz) and wide-band (7 kHz) conferencing systems. Narrow-band applications include teleconferencing and low bit-rate video conferencing. Wide-band applications include high-quality teleconferencing and video conferencing, as well as distance learning. Wide-band conferencing system users should be particularly interested in using an AEC solution, as it will help them to reap the most benefit from the additional audio capabilities of their systems.

People at the remote end of the transmission path are the primary beneficiaries of an AEC. Installed at the local end, an AEC prevents the echo of the remote person's voice from being returned (echoed) to them through the audio system. People on the same end as the AEC should not notice the AEC if it is doing its job properly and since the person on the far end hears better audio quality, the AEC enables the conversation to flow more smoothly.

Be sure not to confuse acoustic echo with line echo. Line echoes are reflections within the telephone line, rather than echoes from a room or auditorium. They are usually only one or two noticeable reflections from telephone hybrids or impedance mismatches in the line and are usually delayed by less than 32 milliseconds and do not change frequently, if at all. Acoustic echoes have a very complex path with dozens or hundreds of reflections that last 100 to 200 milliseconds and can vary continuously during a conversation.

In order for participants at both ends to hold a full-duplex, hands-free conversation, both ends must be equipped with an AEC as in Figure 4.

TELEPHONE SYSTEM FEEDBACK

If your sound system starts feeding back whenever a phone line is introduced, the problem is not in the sound system but in the interface to the phone line. Most often, the wrong device has been used for bringing the phone call into the audio system. Telephone couplers, which normally cost $300 or less, simply will not work for your application because they cannot adequately isolate the two sides of the telephone call. This inadequate isolation results in bleed through of audio from the send side of the coupler to the receive side. When this audio is amplified through your sound system, electronic feedback results.

Luckily, phone coupler problems are easy to resolve. Throw out the couplers and replace them with digital telephone hybrids. These devices are the same products used at radio and TV stations to bring callers into talk shows. They are considerably more expensive than the couplers, but they won't introduce feedback into your audio system.

There is only one trick to using telephone hybrids: if the audio sent down the telephone line contains any of the caller's audio, feedback will result. The trick is to use what is called a mix-minus feed to the caller, a mix of all of the audio from your system minus the caller's audio. If your mixing system does not have mix-minus capability (most don't), you can either use a separate mixer for the phone line or buy a digital telephone hybrid with automatic mix-minus capability.

CONCLUSION

Here are the top nine things you can do to design good conference systems:

Place one mic within arm's reach for every two to three talkers.
Apply noise cancellation to every mic channel.
Limit user control to loudspeaker volume, with limited range.
Design conservatively when the remote sites' characteristics are unknown.
Design for as big a difference between NAG and PAG as possible (15 to 20 dB is great).
Ensure nominal levels are sent to and received from the codec.
Use acoustical treatment to reduce reverberation.
Provide full-bandwidth program audio to wide bandwidth codecs.
Design a conference room with the same care you would for a sound reinforcement system.

It is hard to achieve good sounding audio in conference room networks. There are many obstacles including room acoustics, electronics and user technique. However, audio problems are not insurmountable. Once your audio system has been correctly tuned, you will find a dramatic increase in your client's productivity and enjoyment.

Michael Pocino is an engineer at ASPI Digital in Atlanta, Georgia. Michael can be reached via e-mail at michael@aspi.com.

What to look for in an automatic mic mixer

When you're looking for an automatic mic mixer, make sure you pick the right one. The mic mixer you choose:

SHOULD be able to handle the number of mics you will be using (most are expandable or can be cascaded).
SHOULD provide an automatic gating threshold. This means that a mic will only turn on when the sound level exceeds a pre-determined level. The smarter mixers can automatically set the gating threshold above the background noise.
SHOULD permit a chair override on one of the channels. This permits the leader or moderator to take control of the mixer by simply speaking into his or her mic.
SHOULD allow one mic to be always on. If all mics turn off, that room's audio will go away (including background noise), making it sound as though the connection were cut off. Leaving one mic on will provide a more natural sound (and is essential for the proper operation of acoustic echo cancellers).
SHOULD NOT require a specific type of mic. Ignoring this can prove costly.