Hello All --
I am brand new here, and nearly new to audio and video. I was a master control engineer years ago (worked my way through EE university), so I do understand some basics. Of course things are all different now! I hope this post doesn't break forum rules. I won't feel bad if it's moved or zapped as long as someone points me in the right direction. OK...
I have been producing promotional and training videos for my astronomy automation software business. [="http://www.youtube.com/watch?v=UPkXCg_HVbQ"]Here is an example[/]="http://www.youtube…"]Here is an example[/]. Despite using a decent AudioTechnica 2020 mic, I'm unhappy with the audio, so I have gone on a mission to learn how to fix up the dry audio and make it more professional sounding. This material is not going on to broadcast. It will be distributed via HD on YouTube (like the example above). Also, it will almost always be played on earbuds or computer speaker systems, rarely (if ever) on anything resembling a home theater system!
I have achieved some results, but my uneducated ear needs calibration from some experts. That's why I came here! I wonder if some of you would be OK with helping me by listening to a dry sample and then three processed samples and giving me feedback. The room has a couple of obvious resonances which I tried to minimize. I plan to try treating the room within the limits of my better half :-) Also there is some background noise from a washing machine which was a "fact of life" this afternoon while making these test clips. It won't be running when I do the real thing.
[[url=http://="http://soundcloud.c… Here are the four 60-second samples on SoundCloud[/]="http://soundcloud.c… Here are the four 60-second samples on SoundCloud[/]. I think it has blurred the differences, but then I assume YouTube will as well, so I thought this would work. I could upload them as little HD videos if you would rather (and I probably should do that anyway).
I just added a fifth track to the set. Thanks again!!!
Start with a good microphone a good preamp and a good ADC. Get one of these [="http://www.seelectronics.com/reflexion-filter-pro"]sE Electronics Reflexion Filter Pro: The Original Portable Vocal Booth[/]="http://www.seelectr…"]sE Electronics Reflexion Filter Pro: The Original Portable Vocal Booth[/] or [[url=http://="http://www.primacou…"]Primacoustic Acoustic Solutions[/]="http://www.primacou…"]Primacoustic Acoustic Solutions[/]. Unfortunately the same thing applies in audio as in computers "garbage in garbage out" If you don't start with a good clean signal all the processing in the world will not fix it. FWIW and IMHO
Thanks for your reply and the suggestions for killing the room resonances. Those look good, and I didn't know about them.
Did you listen to the samples? I listened to maybe 100 samples of audiobook reading from my own audiobooks (at max resolution) and from the large collection on SoundCloud. Most, even those which are professionally produced, are IMO pretty bad. One expensive audiobook of mine was read in a booth with such a prominent resonance that it actually made it difficult to understand the reader. I didn't post here until I had produced some tests that sounded better than most (at least to me, and that's the point!).
I'm not trying to be argumentative... If you think the samples are really that bad, please do say so and I'll start over using your suggestions as the starting place. I have thick skin so forthright criticism, specific criticism would be welcome. I need to know what's wrong with what I've done in order to do better.
I like the dry signal and #4 the most the other sound "funny" My company, Acoustik Musik, Ltd., does a lot of voice over work. We do commercials, voice over training recording for a company in New York plus we have done work for the Amercian Symphony Orchestra and for the History channel. We also have produced two nationally syndicated radio shows so I am familar with good voice over work. We also have done a number of talking books. Our setup is as follows,
One Shure SM-7B Microphone
One [="http://www.seelectronics.com/reflexion-filter-pro"]sE Electronics Reflexion Filter Pro: The Original Portable Vocal Booth[/]="http://www.seelectr…"]sE Electronics Reflexion Filter Pro: The Original Portable Vocal Booth[/]
One Robbie Microphone preamp
One Benchmark ADC-1
Custom made baffles and microphone stand filled with sand for anti-resonance
Room designed by DSM and Associates [[url=http://="http://www.dsmassoc…"][ DSM & Associates, LCC ] Consultants in: Acoustics, Sound Systems, Audio/Visual Systems, and Recording/Production/Broadcast Facility Design[/]="http://www.dsmassoc…"][ DSM & Associates, LCC ] Consultants in: Acoustics, Sound Systems, Audio/Visual Systems, and Recording/Production/Broadcast Facility Design[/]
Your stuff sounds good. But it could sound better. Again post processing is icing on the cake and not making a cake from scratch.
Best of luck.
You can call me if you want to talk.
Thank you so much! I just needed a sanity check. I also like #4 . It was a "last minute" experiment so it was added after the others. I have looked at the two sound control things you posted. Thanks again for the pointers to them! I have to trade off "that last 10%" against turning my office and studio into a sound booth :-) so those portable devices look like the ticket. I completely understand that the best solution is to start with good acoustics instead of trying to notch out resonances after the fact.
Perhaps a few others will weigh in here. Again I am all ears (ha ha, well -untrained- ears that is).
I've done an awful lot of commercial & voiceover recording along with 20 years spent at NBC-TV doing news and political talk shows. While Thomas makes some good suggestions, there is absolutely nothing wrong or inferior about what you are currently using. As an electrical engineer you understand certain concepts of design philosophies. Recording audio isn't any different. It's more about how you go about it than the equipment in the equation. You're not trying to produce a hit record. You need spoken word recordings that are pleasant to listening to, sound full and robust along with being authoritative. You need proper equalization, dynamic range modifications. I do hear a reflection in your current examples. It sounds like you are sitting and facing a wall about 3 feet away from you. It also appears you may have a script directly in front of you, behind the microphone? This is causing you more acoustical aberrations that are generally peculiar and uncomfortable to listen to for any length of time.
Well we all know your microphone is decent. It's not great nor may it be the proper microphone to be utilizing in your surrounding acoustical environment. Spoken word for your application should have a intimacy to it. The only intimacy I hear are 2 people trying to squeeze into a single telephone booth. Some of these acoustic problems can be remedied by simply repositioning. For instance, you don't necessarily want to speak towards a reflective surface such as a wall. You want to speak into a larger unobstructed room space where reflections might be further away from the pickup to the microphone. You should also be 3-6 inches from the diaphragm of the microphone. You will also likely want to include a high pass filter around 80 Hz to lessen the " Proximity Effect ". Now listening to it in its raw state like that you may not be pleased with. So on to the more fun stuff...
Here is where some dynamic range limiting and/or compression comes into play. You want anywhere from 2-10 DB of gain reduction at around a ratio of 4:1. Attack times should be set modestly to retain some dynamic inflection of your voice. On the release time side, slower or sounds more natural while faster creates more of a sense of more "apparent loudness", While increasing the aggressive nature of the sound. That's not necessarily what you want for astronomical descriptions where it is more appropriate for fast-paced, hard-hitting commercials & promotional material recording purposes. But this also helps to all exaggerate and accentuate your lousy acoustic surroundings along with increasing all of the background noises nobody should be hearing. Did I hear you fart? We don't want to hear that either. So after your dynamic range compression and/or limiting, you need some downward expansion. Downward expansion is similar to gating. A gate is a popular device to utilize on things like snare drums and bass drums. That's where you want the microphone to be off until the drum is hit. And then it is only on for that split second the drum is being hit and is immediately gated off again. That's fine for that application. It's simply horrific sounding on spoken word. It's completely unnatural sounding. Downward expansion only differs in that the amount or, DB's a gating can be restricted to a preset level such as only -10 DB instead of completely off as a gate would do. Threshold levels are very critical in this adjustment. At that time should be as fast as possible but release time should be adjusted for the most natural and unnoticeable lowering of the gain possible. This can be accomplished with the most audio software packages though, I've not found any yet whose presets are anything other than a hard slamming gate. So one must utilize the standard broadband compressor most software packages have. These types show a graph of a typical compressor gain modification curve i.e. brickwall limiting, soft knee compressor. You would then set a DOT between your softest speech and where your breath is. That establishes your threshold. As you will see, below that DOT the line will travel in a direct linear path to zero level. It is there you will place another DOT and drag it slightly to the right, on the baseline at the bottom. This will provide some smooth downward expansion. When done properly, it's not only an effective use of ambient noise reduction, it will also eliminate the gasping effect of breathing that the compression has accentuated. So you get a nicely processed and tight sounding vocal with natural breaths and a nearly total elimination of acoustical room problems.
I spent a few minutes with this to give you an example of the above. This is from your first sample as a base reference. This is rough but it should give you a better idea.
Personally I think you would do better with a SHURE SM58 or, Beta 58 in your current surroundings. They are generally less sensitive to outside extraneous noise influences and have a slight bit of bandwidth limiting in comparison to a condenser microphone sound.
[[url=http://[/URL]="http://soundcloud.c…"]Bob Denny 1 by user3139903 on SoundCloud - Create, record and share your sounds for free[/]="http://soundcloud.c…"]Bob Denny 1 by user3139903 on SoundCloud - Create, record and share your sounds for free[/]
Mx. Remy Ann David
Thanks very very much! This is info I can really use. You are right on about having a script in front of me, The (side-address cardoid condenser) mic was above my mouth near eye level and I was looking below it to the script (actually a passage from Watson's book on discovering DNA). I know that a cardoid mic picks up stuff from the "null" side. Worse, there are two big-ass computer screens right in front of me. Now I know what's making the audio sound like it does. It makes sense.
I have a chain consisting of parametric EQ, single-band dynamics, a touch of harmonic excitation (subtle to my novice ear), and a touch of stereoization (on #4) which seemed to make a huge difference on monitors and especially on earbuds/headphones! It amazed me that only a few degrees spread on the vectorscope could make that kind of difference. Thomas' "funny" sound is a result of notches to kill resonances. The one at 1720 is the most obnoxious. The info on how to set up the dynamics is great. I get it. I know what downward expansion is. Guess what? I do have it virtually gated in my clips! Score another for the experts! It's actually a 4:1 downward at -46dB, 60ms attack, 300ms release. By the way I found the resonances using the parametric EQ by dragging a high-Q peak around until I hit a "hot spot" where I could hear the echo and resonance, then pulling it down into a notch. Repeat three times. Ugly.
I super appreciate your doing your own number on the dry clip. It's just that sort of reference that can help me build my experience. I couldn't have asked for better input. THANK YOU BOTH!!!!!
I'll do some things, work on the processing and be back with new results.
You simply need a little more compression and a better setting of the threshold on your downward expansion. It's always difficult dealing with scripts and music on music stands. I made an operatic recording of a soprano for Delos records. She was a fabulous Wagnerian soprano with approximately a 1 hp voice which equates to about 750 W. I utilized a AKG 414 B-ULS on her. Her music stand of course was at a slight angle to her. But I was getting a peculiar audible kick back from her music. We adjusted things numerous ways and so did she. Nevertheless, this still caused phase abnormalities. I finally got her to lower her music stand and angle it completely flat and parallel with the floor. It was only then that it started sounding quite a bit better. And this was in a beautiful huge cathedral in Wellington New Zealand with the New Zealand Symphony Orchestra accompanying. I was going to utilize my Beyer M-160 ribbon microphone on her. But due to a logistical problem, I wasn't able to use my API 3124's on this session. Instead, I was presented with a AMEK BC 1 mixer. It was smooth sounding. Too smooth to be utilized with that ribbon microphone and that's why I switched to the 414. I had recorded this singer before and really preferred the ribbon microphone on her as long as I was utilizing my API's. It's OK, I still received a Grammy nomination for best engineered.
Can't wait to hear what you come up with now.
Mx. Remy Ann David
Again, thanks to you and Thomas too. It's going to be at least a few days before I'll have the time to work on this again. I have some ideas on how to minimize the room acoustic problems now, but that will take longer. I will definitely post back here.
I really appreciate your help and advice!!!
Apologies to those who took the time to help me. The realities of my work (astronomy software) have intervened and prevented me from continuing toward my goal of new videos. It looks like it will be a week or more additional. I just didn't want you all to think I went away unhappy of whatever. I do appreciate the help and advice!!
Remy and friends --
Here it is May! This is a 2-person business and opportunities have resulted in my priorities being shifted around such that my videos have taken a back seat. Gotta make the money when the opportunity arises :-) Once again I sincerely appreciate the help. I'll be back when I've cleared the decks and am again ready to tackle the videos (and of course the narration audio that they need). I just didn't want you to think I gave up or something.
LOL Bob, we've all got to do what we've all got to do. It was only yesterday I was cleaning out the files on my computer when I ran across those earlier narrations I had helped you with. Wow, what a coincidence. Keep us posted, will you?
It's like déjà vu all over again
Mx. Remy Ann David