The disadvantages are also potentially significant:

• Not everyone is confident using a computer microphone over a traditional phone.
• VoIP is completely dependant on the quality of your network, if the network fails, the audio fails.
• Computer audio invariably needs to deal with feedback and echo cancellation.

For the purposes of this post, we will discuss a purely VoIP environment.

It is good to understand what is happening in Connect when you use VoIP if you want to be able to troubleshoot your sessions.  The key to a quality experience is understanding echo cancellation, also known for the purposes of this post as ‘feedback’.

If there was no echo cancellation in Connect, when you said ‘hello’ through your microphone, your computer speakers would play back ‘hello’ immediately afterwards, which would be ‘heard’ by your microphone and then played back through the speakers again and again until you have that dreadful whine that scares everyone.

Echo cancellation is the tool inside of Connect that says “Ian just said ‘hello’, do not say ‘hello’ back through Ian’s speakers, just say ‘hello’ through everyone else’s speakers”.  When this is working correctly, you will not get echo or feedback.

Echo cancellation is less necessary if you are using headset, simply because your microphone cannot usually ‘hear’ your headphones, whereas a camera microphone can easily ‘hear’ the big speakers on your desk.

It is also fair to say that a headset offers a far better quality of microphone in that is is attached to your head, so it will ‘follow’ your movements.  The microphone on a headset is also very likely to be noise cancelling and it will filter out background noise and just reproduce the sound of your voice.

Echo cancellation is still in play with a headset however, as we will discuss when we look at mobile devices later in this post.

Best Practice – If you can use a noise cancelling USB headset, you will have the best quality audio and less challenges with feedback and background noise.

In many cases where you see issues with feedback during your tests, it is actually more of a function of your testing environment than a result of Connect not performing.

Here is the scenario to avoid – You decide you want to test your VoIP audio and so you set up two computers.  You start VoIP and turn on the microphone and hear the most awful feedback imaginable.  In this case the echo cancellation is working, but having the two computers side-by-side is negating the tool.

Using the same analogy as earlier, you say ‘hello’ into the microphone on computer #1, so computer #1 knows not the say ‘hello’ from its speakers, echo cancellation is working as expected.  The issue is that you have computer #2 sitting there too, and computer #2 happily repeats your ‘hello’, which computer #1 hears in its microphone and repeats on computer #2 again…hence the feedback.

The echo cancellation is designed to work on the computer you are using, if you have multiple computers sitting together then they ‘listen’ to each other and this bypasses the echo cancellation.

If you think this through, it should not be an issue as the whole point of VoIP is that people are in remote locations, so this should never be an issue in a real-life scenario.  It is a common issue during testing however.

Best Practice – Test with people in separate locations to mimic as best you can the real-life scenario for Connect.

Mobile devices such as iPads, iPhones, Android Phones or tablets use the ‘Connect Mobile’ App to attend a Connect Meeting as there is no native Flash Player on these devices.  The App is very sophisticated but there is no echo cancellation (or at least it is so poor  as to be negligible).

If you are using a mobile device and you want to use 2 way communication you will absolutely need to use a headset to avoid feedback, or enable ‘Single Speaker Mode’ and even then you have to be careful with volume as we will discuss here.

Note – This is not how this should be of course, it is an issue with the Connect Mobile client that should be fixed.  The intention here is to make the best of what we have today.

In this scenario, you do not really have to worry about echo cancellation or feedback as you are not going to enable the microphones of the participants, hence there is no opportunity feedback.  If you only have one speaker then you can just broadcast from your microphone and nothing should be coming back the other way to interfere with your audio.  

If you want to be totally sure that you are not going to hear feedback, then use the ”Enable Single Speaker Mode’ on the ‘Audio’ menu in the meeting room.  This will ensure that only one person can talk at a time and you can see that is is enabled by the small asterisk beside the microphone in your meeting room.

If you are broadcasting from a mobile device you can use the inbuilt microphone or a headset, as there is no two-way communication it really does not matter.  We would recommend that you use a PC for the meeting lead though, Connect will allow you to lead a session from and iPad of an Android tablet, but there are far more controls in the PC version and it is a far more sophisticated implementation of Connect.

Best Practice – Always use a PC over a mobile device if you have an option, particularly if you are leading the session or presenting.

This scenario has the same considerations as scenario #1, with the exception that you need to consider feedback across presenters in the same room and feedback or background noise form other presenters remotely located.

The most water-tight option is to use the ‘Enable Single Speaker Mode’ outlined above.  This is particularly appropriate if the presenters are going to speak one after the other as part of a more formal agenda.  Speaker #1 presents, then speaker #2 presents etc.

With ‘Enable Single Speaker Mode’ engaged, speaker #1 turns on their microphone and no other speaker can use their microphone until speaker #1 releases it.  This ensures that only one person is speaking and as such there is no opportunity for feedback or background noise.

As the Host of the meeting, you will also have the option to force the presenter to release the microphone when they are done by selecting ‘Release Mic From Speaker’ from the ‘Audio’ menu in the meeting room.  Then you can ask the next presenter to enable their microphone.

Best Practice – Use ‘Enable Single Speaker Mode’ whenever you can.

In this scenario you might be having a panel discussion or open Q&A session at the end of a set of presentations (after Scenario #2 above for example).  You could use ‘Enable Single Speaker Mode’ but it will make the discussion rather stilted and disconnected as each person has to release the mic and enable their mic to speak.  In some case this may be something the meeting Host prefers to control so ‘Enable Single Speaker Mode’ may be a good option if the meeting Host uses the “here is a question for Dr. Jones” process.

A more likely scenario is that you want an open and interactive panel discussion so do not use ‘Enable Single Speaker Mode’ in this case, you can enable the microphones for all of the presenters but consider some best practices to mitigate any issues:

• Ask the presenters on a PC to mute their microphone when they are not speaking, just click on the microphone icon and a line through the microphone will appear when it is muted, un-mute in the same way.  This avoids background noise and mitigates any feedback (although this should not be an issue with a PC).
• Ask presenters on a mobile device to disconnect their microphones when they are not speaking, their is no ‘mute’ on the mobile devices, just ‘connect’ or ‘disconnect’.
• If the presenters are looking to leave their microphones on the whole time, ask them to be conscious of background noise, it can be very distracting.  As a Host, keep an eye on the microphones in the Attendee List Pod as you can often see the background noise and identify and mute the offending presenter (mouse over the name and then ‘Mute Attendee’).
• If you hear feedback then you can identify this in a similar manner by watching the microphone icon in the Attendee List Pod, when you think you have the offending presenter, mute their mic and see if this fixes the issue.
• Remember that Connect Mobile has very limited echo cancellation capabilities, so a headphone is essential in multi-way open-microphone conversations, but even with a headset you can get feedback.  The microphone on a set of Apple Ear Pods for example, is only a couple of inches from the ear-speaker.  If the ear speaker is set to a high volume, the microphone can still pick up the audio and you will hear a slight echo of your voice. In these cases, the best practice is to ask your presenters to turn their headsets down as low as is reasonable for them to hear the audio.  The louder the audio in any scenario, the more chance there is of echo or feedback.

Best Practice – Have PC-based presenters mute their microphones when not speaking and have Connect Mobile-based presenters disconnect their microphones when not speaking.

This scenario is really not very different from Scenario #3 in that you have multiple speakers.  The best practices above all apply but consider that this can get out of hand in a hurry.

If you have even as few as 10 people in a meeting room and you enable the microphones for all attendees and leave them ‘open’, there is a fair chance that you will always be hearing a dog barking, a doorbell, a sneeze or any amount of noise from somebody on the call.  Nothing to do with technology, this is a result of human nature.

In a physical environment, you would not typically leave it open to let people shout out whenever they want, whenever they want, you would typically ask a question and then choose the hand raised to answer.  Just be conscious that you might want to mimic physical best practices in the virtual environment.

That said, you may have open conversations with multiple open-microphones, so apply the best practices we have discussed in Scenario #3 for the best experience.

Best Practice – If you would not do it in a traditional physical environment, you do not need to do it in a virtual environment.

Most sessions are a mix of the scenarios above.  You might start a company meeting with presenters one after the other and then move on to a panel discussion.  You might run a class by being the single speaker initially as you teach and then move to a class-wide discussion.

If you can identify these phases in the session and apply the practices above consciously then this will help you mitigate the risk of issues with audio.  One size does not necessarily fit all.

Best Practice – Identify the phases of a session and apply the appropriate practices to each scenario.

As described above, there are some ways in which you can troubleshoot audio in Connect, something will often go awry and in some cases there is absolutely nothing you can do, but there are a few practices that you can apply:

• Prepare Hardware – On a PC, have presenters use the ‘Audio Setup Wizard’ in the meeting room to ensure their microphone and speakers are set up correctly.
• Help your Users – Tell mobile users to have a headphone available, tell them to lower volume as far as possible etc.
• Know Connect – As a Host, become familiar with the audio option in Connect and how to ‘Enable Single Speaker Mode’, mute people, enable microphones, disable microphones etc.
• Know your Troubleshooting – As a Host, learn how to identify problem users and mute them, isolate issues etc. in the Attendee List Pod.
• Break Something! – As a Host, set up some tests where the best practices are *not* followed, generate feedback on a mobile device and work out how to fix it.