What do you recommend changing in my audio system?

[Adriano Moroni]

Hi, I have no experience, but I have good ears. ;) I use the sound system to watch my documentaries. I do the video editing myself and there is a narrative voice in all of them. My audio system includes 2 tweeters, a central speaker (for the voice), a subfoofer and 2 rear speakers (which cannot be seen in the photo). I've marked with arrows the speakers I'm using and the Onkyo amplifier.
My Bose audio system is 20 years old but I replaced the central speaker about 10 years ago and it cost me 100 euros.
Unfortunately when I listen to my documentaries, the voice is not good. Yet if I listen to it with 2 cheap studio speakers, I hear an excellent voice because it is recorded with a mic. professional and by a professional speaker.
I had thought about removing ONLY the central speaker and buying two studio speakers and leaving everything else unchanged. What do you think. Maybe I said something rubbish, but I'm only interested in hearing a good voice mixed with background music, but with this system I don't hear it well. But I'm satisfied with the music that comes out. What do you recommend me to do? I have a limited budget. Thanks for some info.

[Andrew Smith]

If you are having difficulty with discerning spoken words, the first thing I would do is to make changes to the room to attenuate any sound reflections (echo / reverb type stuff).

Curtains and other furnishings may help with absorbing noise ad minimizing the return echo within the room.

Andrew

[Bruce Watson]

Originally Posted by Adriano Moroni

Unfortunately when I listen to my documentaries, the voice is not good. Yet if I listen to it with 2 cheap studio speakers, I hear an excellent voice because it is recorded with a mic. professional and by a professional speaker.

First, is the dialog only on the center channel, or is it also on the L/R channels as well? Are these documentaries some that you own on disc, or some that you have created and therefore have the masters? Knowing the answers to these questions might help people give you more direction.

From the picture, I see hard tile flooring and a small rug. This implies that your listening room generates a lot of audio reflections (as hard surfaces do). This in turn can be bad for "intelligibility" of the voice channel(s). Small rooms tend to generate rapid first reflections (commonly called "slap echo") which tends to blur the direct sound and create a sort of "smearing" of the audio. It makes it hard to localize sources in a stereo sound field for example, and tends to smear the consonants in dialog.

Small rooms generally need some level of treatment. Typically small rooms need some amount of absorption, some (different) amount of diffusion, and some bass traps. If you use nothing but absorption you end up with a dead sounding room. If you don't use any bass traps, you end up fighting room modes (nulls and peaks at certain frequencies below, say, 350 Hz). If you use too much diffusion you end up with a room that's too bright and too lively.

So, how to know how much is enough, and how much is too much? You need a computer program like REW (room equalization wizard), a calibrated microphone, and some cables so the computer can play tones through your AVR. It works by the computer playing frequency sweeps and listening to how your room responds through the microphone (at your main listening position). You can then generate graphs on your computer that show you what's actually happening in your room at all the pertinent frequencies. As you add room treatments, you rerun REW and see how what you've added changes the room's response.

Or you know it could be something as simple as your center channel being out of phase with your L/R pair. REW will tell you this too I believe. But so will swapping the ends of the cable at the center channel.

Once you have calmed your room's responses with treatments, you can use a modern AVR to run another software program like Audyssey multeq xt32 which can fine tune your system response to really very flat indeed. But programs like Audyssey, Dirac, ARC, and others are built into AVRs and thus require you to upgrade your AVR to make use of them.

This whole thing of making a room sound good is a large endeavor. Most people seem to do just enough to cure the thing that was making them unable to listen like they wanted, and just stop there. Really, you don't have to make your listening room into a recording studio. And the treatments can be as simple as a bigger rug to cover up more of those tiles, and more furniture in the room, say a mix of upholstered (aka absorption) pieces and things like book shelves with a fair amount of books (aka diffusion).

Well, I've given you enough search terms you can probably find more in-depth discussions as you need them.

[Adriano Moroni]

I thank you very much for your interesting reply.
I need you explain better. Near that small carpet there is a bigger carpet, then there are the curtains. It's not a soundproof room but it's not bad.

I make my documentaries and I make the video editing. I have the master for every documentary.

I'm thinking to buy one of two speakers Elac DEBUT 2.0 B6.2 or Elac Debut b5.2 and plug it on Central. I will remove the 2 small tweeter. For me the best thing is to listen a good narrative voice in my dicumentaries.
Since I also have a Subwoofer (look at the photo), I think I'll buy Elac Debut b5.2 instead of Elac DEBUT 2.0 B6.2 is enough to listen a good voice. What is your thought?
Thank you again.

[Bruce Watson]

If dialog is isolated to the center speaker (as is normally done in 5.1 soundtracks), then the problem is more likely to be the room, or the speaker. If the dialog is spread across the front (L/C/R) as is sometimes done so that dialog can be heard in situations where a center speaker is not available (using the L/R pair to create a "phantom center channel"), your problem could be caused by one or two speakers being out of phase with the others. This will make the dialog sound "hollow" and it will be hard to locate in the room (it will not sound like it's coming from the center speaker as it should). Flipping one end of the speaker wire coming into the various speakers will eventually reveal a fix for that problem too.

So in the first case, a new center channel speaker may help. Elac has a range of center channel speakers, I would tend to use one of those instead of one of their "bookshelf" speakers because of the improved horizontal dispersion, which helps intelligibility. I'd probably push you more toward the Elac Debut Reference Center Speaker – DCR52 (or the less expensive ELAC Debut 2.0 C6.2 Center Channel Speaker – DC62), if it fits where you need it to go. But of course, it's your decision. Whatever you buy, set it up so that it points directly at your ears (you may need to tilt it up or down a bit, as well as pointing it L/R) when you are in your main listening position. This gives you optimal direct sound (maximizes signal to noise (reflections) ratio).

In the second case, dropping the dialog from the L/R pair and letting the center channel carry the load alone will probably help quite a bit. Reflections in a small room are tricky; I like my reflections to have a single source when possible. When L/C/R are all the same signal (but spaced apart) they create three slightly different sets of reflections; things get messy quickly.

[Adriano Moroni]

Your reply clear many many concepts, but perhaps I have too many gaps. I'm not an expert.
The dialog of my documentaries is always isolated to the center speaker. But I listen a bad narrator voice.
Maybe that center speaker I have (JBL) doesn't communicate well with the 2 very old Bose tweeters. I don't know this, but I am dissatisfied. So I don't think I'll leave the dialogue of my documentaries only in the central speaker. Perhaps by adding 2 more Elac speakers on the Centre, listening could improve. I hope!!

[Bruce Watson]

Originally Posted by Adriano Moroni

I make my documentaries and I make the video editing. I have the master for every documentary.

I watched one of your documentaries for awhile:

WAYUU a desert tribe of Colombia with amazing traditions - Guajira - YouTube

I am listening on a pair of headphones, Sennheiser HD650s. I don't have actual speakers in my office to compare to -- my wife would object to my noise making ;-)

I think I can hear at least some of what you are complaining about. Starting about 5:08, the interview with the man in the blue shirt. When he is speaking (and is up in the mix where we can easily hear him) he sounds nice and clear. In comparison, I find the narration somewhat muffled. I think this is primarily some amount of roll off of frequencies above, say, 5 kHz. I could easily be wrong, but I think that's what I'm hearing. Consonants, mouth clicks, etc. run up to around 10 kHz; these sounds are clues to the brain that help it translate what the ears are sending to it into phonemes, the parts of human speech that the brain stitches together to get words. I don't hear this lack in the field recorded interview, only in the voice-over narration. I'm wondering if this is part of what you are complaining about.

[Adriano Moroni]

Yes, the problem is the voice-over narration. But the voice of the doc. that you have listened satisfies me. Now I no longer have that speaker, but I have another one with a more powerful and bigger voice. Now I even struggle to understand the words if the music in the background of the voice is not mixed perfectly. Listen to one of my docs. it got a little stressful because of that.
For this reason I would like to do something to my audio system, but without spending a lot.

[Rick Reineke]

Some of the narration sounds ok and some not so good. From what I hear on my reference monitors, most of the music is significantly louder than the narration.
Are you mixing in 5:1 or Stereo,

[Adriano Moroni]

Stereo only.

[Bruce Watson]

Originally Posted by Adriano Moroni

Yes, the problem is the voice-over narration. But the voice of the doc. that you have listened satisfies me. Now I no longer have that speaker, but I have another one with a more powerful and bigger voice. Now I even struggle to understand the words if the music in the background of the voice is not mixed perfectly. Listen to one of my docs. it got a little stressful because of that.
For this reason I would like to do something to my audio system, but without spending a lot.

I understand. But I don't think the problem is your audio system. I think it's your signal.

If you've got the masters, you must have the voice over-track. Pull that track into a DAW (digital audio workstation) program and look at the spectrum for the track. You'll probably have to use a shelving EQ on the top end to bring back the attacks on the consonants and the "air" as they call it; the spectrum will tell you where to start and show you how it flattens as you raise the top end. You don't have to make it ruler flat, it should probably slowly roll off above 10-12 kHz -- maybe be 3dB down at 15 kHz? IDK, it's been a long time since I did stuff like this.

A dialog editor can tell you more about what else to do, but this particular voice sounds a bit flat to me so it might need some EQ boost in the range around 2.5 kHz. If you have a parametric equalizer available, use a fairly low value for "Q" to make the bandwidth of the correction more broad and less of a peak. Use only as much boost as you have to so you don't leave any audible artifacts. Maybe 2-3 dB of boost I'm guessing. Maybe a smaller boost (similar) in the low end in the 150-300 Hz range to give him a touch more warmth.

When you get done, swap the corrected track for the old track in your documentary and watch / listen again. Can't hurt to try, and it might save you the time, trouble, and money of buying new speakers.

Buona fortuna, spero di essere stato utile.

Oh yes, you said the mix was stereo only. A new center channel speaker won't help you much at all since the L/R stereo speaker pair are providing a "phantom center" as it is. This is part of your problem (isolating dialog to a mono center channel always sounds better, more solid, than a phantom center can IMHO). But for an independent documentary (which you do quite well) that shows at all kinds of locations (where the sound system is unknown in advance) stereo is probably the safest way to do it. I learned that from guys like you years ago. ;-)

[Greg Miller]

A very interesting video!!!

I think there are really two separate issues: (1.) how should the playback system be changed, and (2.) how can the mix be improved?

** Comments about the playback system:

• All three speakers should be as close as possible to ear level and the same level above the floor. They should be centered (left to right) relative to your listening/mixing position, so the sound from all three arrives at your ears at approximately the same time and volume. Where the heck is your center speaker? If we can't see it, it's certainly going to sound different from the other two ... probably muffled if it's hidden behind something. You want the three of them to sound as similar as possible; you do not want some "special sounding" speaker for the center channel.

• You definitely need to be sure all three speakers are in phase. Be sure the amps are connected identically. If your speaker wires are color-coded (e.g. red and black) this is a quick way to check, but listening tests are of course the final goal.

• Is your playback system equalized at all? How? Where in the chain?

• The narration sounds better on your small, cheap speakers because they reproduce less of the frequencies outside of the voice range, so they will make the voice stand out from the other audio. This makes me suspect that there is a problem with the mix.

• I do agree with other folks' comments about adding some acoustic absorption to the room to lessen reflections. That will help intelligibility *if* the voice is loud enough and properly equalized to begin with.

** Comments about the mix: (I listened to parts of the same video that was referenced by Bruce Watson, so other mixes in other videos might be different.)

• I agree with Mr. Watson's comments about muddiness of narration. In fact a big "AMEN" to all of his detailed suggestions (which he wrote while I was writing this). The narration need some restorative EQ, plus careful gain-riding and/or some gentle compression to keep the levels up.

• At times the narrator speaks in a very "intimate" and "personal" tone of voice, rather than projecting his voice at "normal" level. The intimate tone of voice has a different balance between high and low frequencies, so it sounds even muddier.

• In some spots the voice level is definitely lower than the music level. Listen to how the voice trails off at the end of sentences, for example around 6:40, 6:46, 6:55 (in the latter instance he's whispering). The diction is unclear and the voice level is lower than the music level; of course the narration will be difficult to hear.

• Lots of wind noise in the location audio around 10:20 ... a foam or furry wind screen, and/or HPF would help correct that.

• There's an abrupt upward jump in music level at 13:58. (No narration at this point, but why the abrupt change?)

• Contrast is your friend. Just as the white subtitles would be easier to read over a black shadow, rather than over a gray shadow, greater difference between narration and background audio will make the narration easier to hear and understand.

*** Changing your playback system won't automatically "fix the mix" but it should make it easier for you to hear details like these initially.

[Adriano Moroni]

All of you are wonderful people, I don't know how to thank you for your precious suggestions. However, I would like to inform you that my original documentaries have different music and much more interesting than what you hear in the video. I had to replace it because of royalties. So I did another very quick video edit and then put a few minutes of the video on YouTube. The videos you see are much less accurate than the originals.

So I could save money by not buying any other speakers, but by adjusting the position of the ones I already have.
Honestly I was thinking of removing the 2 small speakers because I've had them for more than twenty years and maybe putting in something better I thought it would have been better.

Having said that, I would like to point out a video in which the voice is from the same speaker as this video with which I have listening problems. But both voices were recorded in different studios, but always with mic. professional. The other video is this:

YouTube (Short URL)

Don't you think I could take a test? Instead of raising the central speaker (the one with the voice) I could sit on the ground to understand if anything changes when listening to the voice.
I honestly don't know how to turn up that center speaker, I don't know where to put it to get a better voice. Look at the photo in the first post. Perhaps this is the main problem. If I were to buy 2 more speakers, I would have the space and height to listen better. But then I don't know if it's a waste of money.

[Adriano Moroni]

I would like to send you a few minutes of the voice recording. It's the same file as that voice that I hear badly. The 2° file was modified with the equalizer. Please can you tell me which of 2 files is better in your opinion? I would be very grateful to be able to understand better.

https://www.dropbox.com/scl/fi/bkk9u...joxydz6ri&dl=0
https://www.dropbox.com/scl/fi/1utc7...rnhmei9h1&dl=0

[Adriano Moroni]

Please, excuse my ignorance regarding audio. I would like to ask a question: I noticed that both the narrator's voice and the music come out of the central speaker. This happens whether the narrative voice is present or not. Is all this normal?

PS: Do you think that moving the central speaker to a higher but decentralized position compared to the TV is better? I would like to place it on the left when looking at the photo, but facing the viewer. Look at the photo attached below.Do you think I will hear my voice better?