Total Audacity noob here

Hello all.

Like the subject says, I’m brand new to audio recording/editing and thus, to Audacity (which is an awesome program that I will help support when financially able.) Software devs and their fams need to eat, yeah?

My sole purpose for using Audacity is going to be to record my own voice for voiceover/voice talent work. I’ve been told for years, about once a month or so, that I have a great voice. I’ve never had any formal voice training, per se, but I have been going to the Philip Morris University, Marlboro School for a long time. As I’m closer to 50 years old than 40, I reckon it’s kind of now or never if I want to attempt making some supplemental income utilizing my voice.

The platform I’m currently on is a Win 7 x64 SP1 desktop with a mobo-onboard VIA Envy HD chip. I have v.2.3.0 of Audacity installed. I have a pair of Turtle Beach X12 headphones with a boom mic. So, that is a USB-powered, front-panel 3.5mm audio separate mic input & headphone output plug setup.

When I talk with people via TeamSpeak3 or Discord, they tell me I come through “five by five” or loud & clear, as if I were sitting next to them or on a very clear phone call, with just a bit of background noise. I’ve obviously played with the settings in these two apps to make the outbound and inbound audio sound as clear as possible, with my sound-chip and Win 7 native settings staying at the 50% level.

I’ve tried a few sample recordings with Audacity, exporting them with LAME to .mp3. The volume is maybe half of what I (and others) can hear in TS/Discord, and I can hear some hiss in the background. One of the gents I regularly talk to on TS says he is a TV technician (as in, he works in-studio) and does some successful voice work on the side, using a Rode Podcaster mic, which he strongly advocated I think about purchasing, if I were to be serious about doing voice work. He asked me to send him a raw .wav sample of me speaking and then some silence. That was just about a month ago. He is apparently very busy and hasn’t been able to get around to helping me null-out the 60Hz mains and ambient noise by using notch filters? Is that the right term?

So, I suppose the purpose of this post is to get some general (and perhaps specific) info about where I should think about putting my sound chip and Audacity settings.

And also, I guess I should post a .wav sound sample of nothing but ambient noise?

I almost forgot. Is there an official/semi-official TS or Discord server where real-time advice is available?

I really appreciate any and all help/advice y’all can give to a noob scrub!
-JH

Don’t give up the day-job yet : it’s a crowded marketplace.

Audacity has free plugins to remove 60Hz mains hum & its harmonics , e.g. see …
https ://wiki.audacityteam.org/wiki/Nyquist_Effect_Plug-ins#Hum_Remover

A few seconds of “silence” followed by you saying something.

using a Rode Podcaster mic, which he strongly advocated I think about purchasing

Yes, you need a good mic. Most headset mics are for “communication” and clear-communication isn’t always the same thing as good-natural sound. There are some good (and expensive) headset mics, like the ones Madonna or TV football announcers use, but in the studio it’s usually a cardioid (directional) “large diaphragm condenser mic”.

Rode has a reputation for very-good mics at affordable prices.

If you like to monitor yourself with headphones, there are USB microphones with a headphone jack and zero-latency direct-hardware monitoring. Or if you use a regular studio mic, there are interfaces with direct-hardware monitoring. If you monitor yourself through the computer there is always some latency (delay). The delay can be minimized, but it can be a pain getting acceptable results and IMO it’s best to avoid the issue altogether. (It isn’t always necessary to monitor yourself unless there is background music that you don’t want to be picked-up by the mic.) The Rode NT-USB has a headphone jack.

Another “nice feature” on a USB microphone is an analog recording-volume control.

If you get a regular (analog) stage or studio mic, these are not compatible with regular soundcards. You need an [u]audio interface[/u] with a low-impedance, balanced, XLR connection. The interface will also provide 48V phantom power, which is required for (analog) studio condenser microphones.

Now the hard part… You need a quiet “studio” with minimal sound reflection. There are [u]audiobook standards[/u], including a noise standard, and that’s a good place to start. And, there is an [u]ACX Check plug-in[/u].

If you are doing commercials you’ll probably want a “hotter” and more-compressed sound than an audiobook, but that’s still a good place to start.

And also, I guess I should post a .wav sound sample of nothing but ambient noise?

No, noise alone doesn’t help… You can always turn up-or-down the level… It’s the signal-to-noise ratio that’s important. So, part of the trick is getting close to the microphone and speaking with a clear-strong voice, without getting so close that you damage the sound quality. A [u]pop filter[/u] can help with “plosives” (pops) and it can help you maintain a constant distance from the mic.

but I have been going to the Philip Morris University, Marlboro School for a long time.

What… No Jack Daniel’s?

This is a how-to for producing a forum test. The forum will only allow you to post about 20 seconds of mono (one blue wave).

http://www.kozco.com/tech/audacity/TestClip/Record_A_Clip.html

Everybody misses the first step, two seconds of normal room noise. Hold your breath and Do Not Move for that two seconds. We need to hear just the noises the room is making.

There is a process for audiobook mastering.

https://forum.audacityteam.org/t/audiobook-mastering-version-4/45908/1

And there is a tool to tell you how you did. That’s the ACX Check mentioned in the mastering post.

ACX is a publisher of audiobook readings. As a rule, if you can read for them, you can do anything else. In the recording world, their specifications are not all that strict. It’s usually only a problem if you’re reading from home.

Very few home readers pass background noise the first time. The audiobook specification is background noise at -60dB. In English, that means the room noise has to be a thousand times quieter than your speaking voice. If you can tell your computer is on just by listening, you’re dead. If you have a USB microphone, you can’t move the computer out of the room. The cable limit is about 6 feet (2M).

This recording thing isn’t as easy as the microphone makers lead you to believe.

Koz

Thanks much for the good info.

To summarize: I deleted my audacity.cfg file. What I need to do next is, without adjusting any settings on my sound “card,”, Windows sound settings, or Audacity settings/effects, record two seconds of tomb-like silence, followed by ~15-18 seconds of me talking in a completely conversational tone, with no extra emphasis or over-inflection.

Then attach the file into this thread?

record two seconds of tomb-like silence, followed by ~15-18 seconds of me talking in a completely conversational tone, with no extra emphasis or over-inflection.

Close. Tomb-like only if your room is a tomb. The object is to record your natural background sound. Almost everybody who records this two seconds takes the opportunity to clear their throat, harumph, shuffle in the chair, smooth their shirt, rearrange their script and drop pencils on the floor. You the human should freeze and not make any noise so we can hear your refrigerator, computer fan noise or the jets going overhead. That’s called Room Tone.

You may not make fresh silence with the Audacity tools (Generate > Silence) to cover up a noisy room. We can tell you did that. Also folded in there is the natural noise your microphone makes. They’re not completely dead quiet.

Do the voice in your normal presenting style or the style you hope to use for the actual work. If you have no style (so to speak) then yes, do it in conversational style. ACX’s goal is natural speaking styles, although many performers change style slightly so you can follow multiple characters. If you’re reading a book you will need to present the same way for weeks. ACX hates chapter to chapter changes.

I deleted my audacity.cfg file.

Why did you do that? Was Audacity not working right?

Koz

Well, yeah. He and his partner Jimmy Beam, their good Buddy Weiser, their cousin Busch and a few other fellers too.

No, Audacity seemed to be working fine, except a couple of my trial recordings seemed volume-attenuated, though. I had tried some sample recordings, applied some Effects and played around with the controls while learning the app. I deleted the .cfg to reset everything. Recording and editing audio is a whole new ballgame for me.

James

That’d be kinda creep. though.

Not necessarily. I got insanely lucky. The original people in my house had a son who played drums. They soundproofed the third bedroom. So I moved into a house with a working studio.

I can make almost any microphone work right from high-end pro to floor sweepings. Just walk into the room, set it up and start talking. Environment is not a consideration. I only had one microphone fail and I’m pretty sure it’s because I did something silly. Any day now I’m going to figure out what happened.

a couple of my trial recordings seemed volume-attenuated

Home microphones come out of the plastic wrap like that. For New Users, a little quiet is OK. If you’re going to make a mistake, that’s a good mistake. That’s usually recoverable. If you go too loud and you overload the sound channel (easy to do), then you permanently damage the clip and you have to read it again. No option.

The goal is occasional time line blue tips at about the 50% mark and occasional bouncing sound meters up to about -10dB to -6dB.

You can turn on View > Show Clipping and Audacity will put thin red lines in the blue waves where you got too loud.

If you’re too quiet, that’s not a bed of roses, either. Microphones have a constant hiss (ffffffff) behind your voice. Your job is to make your voice louder than the hiss. If you miss it by a bit, you can use Effect > Noise Reduction to make up the difference. If you miss it by a lot, you read it again because heavy noise reduction can make your voice sound like talking into a wine glass.

If there was ever any question what the recording engineer does…

Koz

More good info above. ^ Thanks.

I’ve talked with a couple of voice pros. The first, an Aussie, said he’d built a portable micro-studio out of thin plywood with soundproofing inside that he can place his head in and record. I asked him if he’d ever worn this contraption out in public, and, if so, how long he’d spent in the clink. Got a good laugh from that.

The second happened to be in a telephone area code near me (let’s say, for argument’s sake, that I live in Texas.) He is either a transplanted British Isles gent or was in character with a voice very smooth, almost sublime, in its quality. He told me that for his first “studio” setup, he used his walk-in closet and pulled a comforter around him to record. He’s since built a dedicated studio. He gave me some other very good advice.

At any rate, my sample clip is attached below. Please rip it to shreds and advise me what to do, where to start, and how best to do it.

Thanks much,
-JH

that he can place his head in and record.

I suspect he meant something like this.



And the object is not to put your head inside as much as it is to announce into it—to reject most of the noise and echo the room is making.

Also here.

A bit better than that is to also put a sheet of something heavy on the wall behind you such as a furniture moving quilt.

579×800 64 KB

That’s the hard way with custom designed wooden sticks. You can also make good use of hardware store plastic pipes. The only restriction of the pipes is not to leave any open ends. Open ends can resonate like a pipe organ.

Listening.

Koz

Cool. I know somebody with a North Caroline version of that voice and he would regularly draw a crowd when he told stories. That’s a darn good start. Sounding like a Hollywood or New York announcer is not required. The goal is to draw a crowd.

I got your submission to pass Audiobook specifications except for noise. It sounds like you’re recording in a cement factory.

Can you tell your computer is on just by listening? If so that’s not a good sign. I think I’m listening to the ventilation fans in the computer since they don’t settle around any standard noise pitches.

Here. I made it worse intentionally so you could get a sense of what’s wrong. It’s actually not this bad.



It’s too loud to fix with Effect > Noise Reduction. The ACX specification is noise at -60dB. That means the noise behind your voice has to be a thousand times quieter than your speech. Can you tell if the refrigerator kicks in? Semi’s or the FedEx truck go by?


The voice level is also low. The ideal blue waves are fatter than that and perfect ones have tips reach up to about half-way (0.5) every so often.

Generally, only the people with full-on sound mixers can make perfect waves, but you should get louder and you can do it by turning the microphone or interface up or get closer. You should be about a Hawaiian Shaka away from the microphone or a power fist if you have a blast filter.

Lets run out of what you got now before you start writing checks. A more expensive microphone might make the background noise much louder and clearer.

Koz

I’m going to do the inventory and see if I have enough half-inch PVC to make a simple sound cave with water pipes.

Koz

There’s a scary test you can do. Can you overload the sound channel—on purpose?

Announce while watching the bouncing sound meters. Keep getting louder and yell if you have to until the meters go all the way up and turn bright red.

If your sound meters are always blue, you may have them set wrong. Right-click the recording meters > Options > Gradient.

Your recording voice should be about half that. So no, you can’t mumble into your beer from across the room.

Koz

This is all great stuff, Koz.
I can’t write or do much ATM as I’m working right now.
The noise is almost certainly from the PC: it sits on my desk about 2 feet to my right front. I have 3 case fans, the PSU fan and a liquid cooling system in it. I built it in December 2012 so the old gray mare is, well, old and gray. I just don’t have the disposable income right now to build a new PC.
When doing my level adjustments, should I take the sound card input up first, say to 75% instead of 50%? Then adjust Audacity?
I’ll do and write more later.
Thanks mucho!
-JH

The level is too low, too quiet. The waveform should look be bigger, like Koz’s picture.
If you can increase the gain controls on the hardware/software that will make it louder.

There is sizzly electronic-noise which is a combination of harmonics from computer-generated 1kHz* signal ,
and harmonics of 60Hz mains-hum. [ * the USB power-supply is usually the source of that ].

Sizzle is harmonics of 1kHz & harmonics of 60Hz.png1086×780 348 KB

Irrespective of the hardware you use, you’ll need to reduce the sibilance: aka De-Ess.
There are free plugins which work in Audacity which do that , e.g. attached.

Steves-de-esser.NY (3 KB)

See … https ://manual.audacityteam.org/man/interface_preferences.html

-96dB is the lowest regular sound can go, so that’s where I left my sound meters. But. Note I also pulled my sound meters all the way across the Audacity window Left to Right. Just click-grab the meter window and pull. The rest of the little windows will scurry out of the way.

-96dB setting is the maximum sound information you can get with the best visibility. You may decide to use one of the other settings if you have a tablet or laptop with smaller screen. I don’t like the -60dB setting because as you found, there’s important stuff just below -60dB and it’s important to know that. There is a compromise meter setting and I don’t remember what it is, but I think it’s in the -70dB range.

Koz

Thank you very much for that information. Adjusted my settings and it’s working a treat, I can see me awake until the early hours now

I’m attaching sample track #2 below. This is with my inputs maxed to (near as I can tell) 100% and using the Hum Removal and Steve’s De-Esser plugins.
Is it normal, when I’m not recording or have any kind of audio program running, to hear myself talking when I’ve got my 'phones on?