[SOLVED] Streaming Audio Challenge

[Penniless Film Maker]

I am new to streaming. I have encountered a challenge in adjusting the loudness of music relative to my microphone.

I am using an Intel 11 Gen desktop system that I assembled this spring with an Nvidia 3000 series GPU. I have no performance issues streaming 1080p with OBS @ 6 mbs, recording locally @ 60 mbs while running Skype and PowerPoint. I have no video games. I do presentations for community groups and schools using PowerPoint.

Last evening I ran a streaming test on YouTube with 10 sites around the country. Each was invited to an unlisted stream via email. I requested each to provide me feedback on impressions.

2 of ten commented that the music was too loud and asked me to reduce the level.

I use Windows Media Player to play royalty free music from YouTube. My selections is thus limited. I use an Elgate Wave 3 run through a noise gate and EQ then into OBS. In OBS I have only two inputs, the Mic and the combined output of all other sources including music from the wave link mixing software. I used side chain compression in OBS to reduce the gain of the music by about 5 db whenever I am speaking on the mic channel. I can hear the music level decrease in my monitor but have excluded the mic from the monitor to avoid confusion while I am presenting.

Several of my reviewers have provided complements about the sound quality and they like the music in the background. I now have to decide what to do about the 2 people who claimed the music was too loud. These people are 1600 miles apart and do not know each other.

It occurs to me that some people may be using high quality HiFi systems while others are on poor sounding laptops and or phones. If these systems have poor bass response my deep voice might sound muddy next to the strong mid-range content in the royalty free music.

I do use an iPad mini to monitor my stream so I know how the music sounds with this small speaker. I normally use a Bose Bluetooth speak with the iPad to improve its sound.

I do not wish to hurt the feelings of anyone who was nice enough to help me with this test but I am reluctant to make changes that might reduce the quality of the experience for those who have higher fidelity systems. I have worked hard the last few months to get the best sound I can from the modest equipment that I have and greatly enjoy the complements that I am getting from most people at the present time.

Is there wisdom available to help with this problem? I will search this week for a less aggressive song. The music I used has no vocals and is rather boring in my view, the next stop is true elevator music with compression to flatten the dynamic range.
Michael

 

[dwd999]

Did you try checking the Windows App Volume and Device Preferences? You have to do this while all of your apps used for streaming are running otherwise they won't appear in the app list. Click Settings, Sound and scroll down to the bottom for the App Volume and Device Preferences. You should see a slider for each app you have running as part of your streaming process. You may be able to adjust the mix between the audio sources to get more pleasing levels. Another possibility is that some listeners are listening in mono and others in stereo so they may perceive the mix differently. It might be best if you converted all your sources to mono to help keep the mix consistent.

 

[Penniless Film Maker]

Hey thanks for this reply. There are some interesting ideas here.

I have several ways of controlling the loudness of the music including the media player, the Elgato wave link mixer and the mixer in OBS. I do use the Windows App Vol but they are just set to 100 and left alone. I monitor the output of OBS in my headphones ( minus my voice) so I could hear the music while I was streaming. I set the level by making recordings and judging the levels while not speaking. I am using compression keyed to my mic so that when I speak the music level goes down by about 5 db and returns to normal about 500 milliseconds after I stop talking. I can hear this modulation both in real time when streaming and in the recordings.

Most people were silent on the issue of music even though I specifically asked for feedback about its loudness. 2 of 10 said it was too loud, 1 said it really enhanced the presentation and the others did not mention it. One person just did not like to song.

I did not think of the stereo issue. The music was in stereo while my mic is in mono. If their polarity was wrong my voice would disappear with the music mostly unaffected. Most of the people involved would not be the type to be using a system where the polarity could be changed.

This week I will use a different song and lower the gain by 3 to 4 db.

 

[dwd999]

I am not sure that having the music level going up and down is a good idea. I would think that once you begin speaking, it might be better to have the music level decrease and remain at that level until after your final word, when you could increase it as a signal that you're finished speaking and the presentation is ending.

 

[Penniless Film Maker]

It is commonly used in the audio industry. It is called "Side chain compression" or "Audio Ducking". It is supposed to be subtle. 5 db may be too much and it does depend upon the type of music. If the music has high dynamic range one should reduce that range using a compressor. Once it is relatively even one can use ducking to increase intelligibility of spoken words without noticeable decreasing the level of the music.

I realize that I am new at this as I just got it working last week. I have to play with the compression ratio and the audio levels and make recording to see how it sounds.

Windows has a function to do this automatically so that its notifications can cut through what ever you are listening to at the time. ( OBS has an advanced audio setting to turn off Windows audio ducking while streaming)

I have to weigh the complements I got from some people against the complaints I got from others. Next week I will try a song with less dynamic range so that I can use softer levels.

 

[dwd999]

That's the whole thing-- you have to figure out what your audience normally likes and cater to them. Just because a lot of producers use audio ducking doesn't mean your audience will like it. It mainly appeals to younger people whose attention span is so short they can't go without constant music. But if your audience is older it becomes a distraction. If you're doing presentations to schools, the kids will love audio ducking but the teachers and administrators, not so much.

 

[Penniless Film Maker]

Interesting observation. For most of my presentations the audience is under 30 while the decision makers are usually well over 30. Most of my presentations in person have no music but when I make documentary file there is almost always music in the background.

When editing video I can move the music up and down as required. This is my first effort where I am automating this function. When I edit video I choose the music first and cut the audio and last I put in the video over the voices of any interviews. This case is quite different.

You have given me much to think about.
Thank You!
Michael