Have I edited my audio well?

Viewing 9 reply threads
  • Author
    Posts
    • #90956
      Cityracer
      Member

      Here is a sample of what my audio for my YouTube videos sound like: https://clyp.it/vhwddw5d. I currently used a Rode Procaster to record it.

      For editing, I use this guide I found on YouTube: https://www.youtube.com/watch?v=9nqCl5gS5iY&list=LLu4V_c-GX55YQjDCp8SczOg&index=1 The only difference is on the noise reduction I use 0.5 sensitivity instead of 0.

      So my question is does my audio sound good? If not, what editing would you suggest instead?

    • #214418
      paulears
      Participant

      what on earth have you done to it? It sounds nothing like the youtube clip. Some syllables are missing, making you sound like kermit the frog, and I cannot understand much of it. Have you applied some weird processing – you seem to have fell into the trap of doing a recording then blasting int with treatment that isn’t needed. Do you have the original, undestroyed version we can listen to to see IF you have problems. I suspect you have hardly any, and have over processed trying to cure faults that aren’t actually there!

    • #214429
      Cityracer
      Member

      I do not have the original audio, but I recorded some new audio examples. Both I will list are unedited. The only difference is one is recorded with the wind screen on the mic, and the other is without the wind screen.
      [With wind screen](https://clyp.it/loggapuz)
      [Without wind screen](https://clyp.it/xqqxo1wt)
      If you tell me which is better, I can provide an edited version.

    • #214444
      paulears
      Participant

      Well – as you don’t have any breath noise evident in either, there’s no real difference. The windscreen seems to make no difference, so I’d stick it on, and forget it. You seem to be quite close and we can hear the inside of your mouth.

      There is a lot of background noise – some kind of air conditioning, people shouting etc, so that needs sorting.

      Now the honest opinion. You don’t have an easy voice to listen to. You might be able to add some eq yo take the roughness off it. People fall into two categories when speaking – those with easy to listen to voices and those not so lucky. You have very little energy in the lower frequencies and most of your volume is concentrates in a very limited bandwidth. It’s a very hard sound and being very honest, I couldn’t listen to it for long. You have some unusual pronunciation on certain words, which could simply be a regional thing, and as a UK listener, I find it tricky. What kind of podcasts will you be doing? I have a friend who has a similar tonal balance to you, and with his UK accent, I find his videos very difficult. Perhaps you could experiment with eq to remove some of the hard edges? Moving the mic away a little could help, but of course then needs the room to be better sounding and controlled.

      Not everyone has a good recording voice. One thing seems clear, you won’t find a magic cure from plug ins or treatment. You could try to re-train your voice to make it rounder, how about finding a singing teacher, experienced in helping people project and control their voices? You don’t want to sing, but many techniques could be the same.

    • #214446
      Cityracer
      Member

      I think part of the issue of my voice not sounding good has to do with a disability I have that makes breathing difficult. When it comes to the eq, I don’t really know what to look for and I’m finding it difficult to figure out what sounds good and what doesn’t.

      I have edited the audio in two ways. One using my normal editing method, and the other using an eq setting called RIAA. Do either sound good, or are they worse?
      [Normal Editing](https://clyp.it/r3c5fffo)
      [RIAA](https://clyp.it/nntlo1f1)

    • #214450
      paulears
      Participant

      bit of both to be honest – RIAA is an eq curve that was used for old fashioned records – these could not have too much bass, because the stylus could jump out of the groove, and the ‘wiggles’ in the plastic would leak into the next groove, giving a pre-echo. So they invented a curve that reduced the bass on record and put it back on replay – so plugging into the phono input on an amp with say a microphone produced a very bassy result. With your voice you need to retain the clarity, but smooth the roughness away. So you need to spend time finding the frequencies in your voice that you can lower in level, leaving the ones that give the clarity. You probably need to look at 450Hz to around 5KHz – and spend some time listening and tweaking. Don’t forget this is how microphones work too – some emphasising certain frequencies in preference to others – so your voice and mic X might just fight, while mic Y doesn’t. EQ does this to a degree, but is a treatment not a solution. However – you need to get your ears trained so you can listen to your voice and somebody else’s and work out for yourself the differences. If you can’t hear it, then you have a problem very difficult to solve. Ear training – NOT random effects. Small tweaks, not huge slap on bass producers.

      Your voice has little at the bottom end, so increasing the bottom end might help – but the RIAA curve is a bit extreme.

    • #214452
      Cityracer
      Member

      I think I’m starting to understand what I’m doing. After much tweaking I have come up with two that seem decent. I did as follows: Noise Reduction, Compressor, Equalization, Normalize.

      [Everything from 500Hz to 5KHz is reduced by 12db](https://clyp.it/0ait1ymo)
      [Everything from 500Hz to 5KHz is reduced by 16db](https://clyp.it/zfbi0fp5)

    • #214462
      paulears
      Participant

      Why noise reduction? The second is very heavily compressed, the first cleaner. What EQ did you use in the end? Personally it sounds very radio compressed now – so it is LOUD – all through, including bits that should perhaps be quieter. All your versions sound different, but not really better? You are treating it like an error being concealed when you just need a bit of eq and maybe a little compression – you seem to just pull settings out of the ether?

    • #214463
      Cityracer
      Member

      I have always done noise reduction and compressor first, because every guide I have read says to. Should I not be doing those things?
      Here are the settings I have used: http://imgur.com/a/pvSQ3

    • #214468
      paulears
      Participant

      I have been producing and being paid for, recordings since 1979, and noise reduction is a REPAIR tool. Compression is also something you use on a case by case basis, to make the audio fit the purpose. This means that for some things, it’s likely I’ll apply compression – but my ears make the decision. For some projects where the end product has a clean wide dynamic range, then I may not use compression at all. For other ones, perhaps with busy mixes where some sources don’t cut through, I’ll apply quite severe compression.

      What is absolutely certain is that any guide that instructs you to remove noise when it does not exist, means you are throwing away something. Like having a rule that says cut all carrots to 6″ long. If all the carrots are 8″ long, then you are throwing away a large amount of perfectly good carrots. Keep in mind that some guides may be written by people who really know their stuff, and others might be written by children at school, who have also read the guides and not realised that context is critical.

      On your settings I see weird stuff. Equalisation is often described as curves – nice gentle starts and ends to the filtering process. You have notches – the kids of eq that is designed for removing complete bands. So 400Hz passes untouched, then you have removed everything at 500Hz – it’s just not nice sounding or musical.

      You must learn to use your ears to do this stuff. Train them to hear subtlety. You are steaming in with very destructive and unpleasant sounding eq.

      If you are using graphic eq, then it’s gentle curves to soften things or enhance.

      Audacity noise reduction as your image shows is there to remove hiss or other things like maybe air con/wind noise – where you can sample a short section of it and then it tries to remove this fingerprint from the recording – it does it pretty well when you have a wreck of a recording, but if your recording is clean – why on earth are you trying to remove noise that isn’t there?

      You are treating this like a painting by numbers exercise. As in do this, do this, do this – and probably before you actually listen to it.

      In your case, you need to do some work to make your voice more smooth. Some people have super smooth voices, where their vocal folds are quite relaxed and can start to vibrate gently, ramping up to full volume on each syllable. Others have more tense and rigid fold that resist the airflow, and suddenly burst into action. Think of a saxophone reed. Beginners start with very thin and pliable ones, that produce tone with moderate airflow. More skilled and experienced players tend to graduate to thicker and stiffer reeds that can be louder, more strident and responsive. Your voice appears to be the second type – so just like a sax, needs some good and effective eq. NOT remotely of the notch variety!

Viewing 9 reply threads
  • The forum ‘Editing’ is closed to new topics and replies.

Best Products

The best video monitors — 2021

We rely on our video monitors to show us an accurate representation of our images throughout the production process. Here are some of the best video monitors currently on the market.
homicide-bootstrap

Need help making a stellar first video?

FREE

Download our free eBook with 8 tips to get on the right track and create a video that you can be proud of.

Giveaway

THE ESSENTIAL VLOGGING TOOLKIT

Given away to one lucky winner

Competition is open worldwide

ENTER TO WIN