Q: How to improve sound and speech recorded on the Voice Memos app in Logic Pro/GarageBand etc.?

[HornetPinata]

Hi all,

 

I've recorded sound on the Voice Memos app on iPhone using the mic on the phone and could really use som advice on how to improve the sound. 

 

I'd like to improve the speech and remove as much of the background noise/hissing as possible.

The recordings are of rather poor quality unfortunately and I was wondering if there was a magic trick or two some of you would know of that would improve them.

I've tried using the EQ and it helps a bit but it's still very much room for improvement.

 

I'm not skilled in using Logic Pro, an absolute amateur in fact, so if there are easier (more idiot proof) ways that you know of to do it in GarageBand for instance (or any other app please) let me know.

 

I'd truly appreciate all and any help as this is important to me. I might end up seeking help from a professional studio but I'd like to give it a go myself before spending a lot of money on this.

 

Thank you in advance.

 

ps. If I take some time to reply please excuse me, I'm more of a night owl on this forum and will probably take some time to answer any replies.

Thank you again.

 

 

[lowdown]

31 minutes ago, HornetPinata said:

Hi all,

 

I've recorded sound on the Voice Memos app on iPhone using the mic on the phone and could really use som advice on how to improve the sound. 

 

I'd like to improve the speech and remove as much of the background noise/hissing as possible.

The recordings are of rather poor quality unfortunately and I was wondering if there was a magic trick or two some of you would know of that would improve them.

I've tried using the EQ and it helps a bit but it's still very much room for improvement.

 

I'm not skilled in using Logic Pro, an absolute amateur in fact, so if there are easier (more idiot proof) ways that you know of to do it in GarageBand for instance (or any other app please) let me know.

 

I'd truly appreciate all and any help as this is important to me. I might end up seeking help from a professional studio but I'd like to give it a go myself before spending a lot of money on this.

 

Thank you in advance.

 

ps. If I take some time to reply please excuse me, I'm more of a night owl on this forum and will probably take some time to answer any replies.

Thank you again.

 

 

 

 

I use 'SpectraLayers' in and out of Nuendo for this type of thing.

My wife does a lot of media voice over work and she's also out and about working on peoples various live Podcasts where there is often back ground noise.

 

I'm not sure Logic has anything like this yet, but Steinberg has a standalone version of 'SpectraLayers'. Also there are various plugins about that do a decent job of audio cleaning.

Otherwise, its a case of getting you hands dirty and chopping into the audio, removing noise between words then using various EQ filters to brighten up the actual audio.

I'm sure there are plenty of YouTube videos using Logic demonstrating Audio/Voice cleaning.

 

Probably any easy way out for you, if you are not skilled in Audio editing, would be to try some of the online 'AI' websites who specialise in cleaning up Audio and voice files. (maybe do some Google searches?).

Usually you can get a few free attempts before committing to purchase of subscriptions.

 

My wife's friend who works on some of the same projects as her uses this company and she is quite happy with the results she gets.

 

LINK:

AI Voice Cleaner & Background Noise Remover

[HornetPinata]

Thank you so much for taking the time to reply so thoroughly and for giving me plenty of options to look into @lowdown

I really appreciate it!

 

I'll look into Steinberg SpectraLayers, and Nuendo, and try to find YouTube videos online. The AI option is tempting to try out as I'm not skilled at all. Used to be able to handle GarageBand but that doesn't cut it here.

 

Thank you!

Edited June 23 by HornetPinata

[lowdown]

6 minutes ago, HornetPinata said:

Thank you so much for taking the time to reply so thoroughly and for giving me plenty of options to look into @lowdown

I really appreciate it!

 

I'll look into Steinberg SpectraLayers, and Nuendo, and try to find YouTube videos online. The AI option is tempting to try out as I'm not skilled at all. Used to be able to handle GarageBand but that doesn't cut it here.

 

Thank you!

  

Sorry, I wasn't advocating using Nuendo, it's just the DAW I use and it was just an example.

Certainly SpectraLayers is worth a look for Standalone use. The new version 11 is excellent.

To be honest if it is something you will be doing on a regular basis is might be more cost effective than going online with AI. I think even the 'Elements' version at £68 should more than cover your needs.

 

What Is New in SpectraLayers 11 | Steinberg

 

Also it might be worth investigating 'RX 11 Elements' by 'iZotope'. It is on sale at the moment.

Both SpectraLayers and RX Elements have a lot of online tutorials and are perfect for content creators for quick fixes with good results as a opposed to in depth audio editing.

 

RX Elements | iZotope

 

 

Good luck and I'm sure others will be along with other options worth considering.

 

 

 

 

[HornetPinata]

6 hours ago, lowdown said:

Sorry, I wasn't advocating using Nuendo, it's just the DAW I use and it was just an example.

Ah. Thanks. As you can probably tell I'm lacking in knowledge here.

I'll have a look at Spectral layers and RX 11 Elements, (and maybe even AI though the whole concept freaks me out a bit).

Having online tutorials to should make it easier to grasp things and as there are quite a few recordings to go through 68 pounds might be a good deal if I get my head around how to use it.

I'll have a look on all your suggestions tonight.

 

Thank you again for all your advice and help.

Edited June 23 by HornetPinata
spelling.

[BigRedX]

How easy it is will very much depend on what sort of noise you are trying to remove.

 

Constant uniform noise is generally the easiest. There is a noise reducer in Logic already, so have a go with that first. Otherwise if there is a section of the recording with just noise you could try copying that onto a new track and then setting it to be out of phase with the original audio and mixing it back in which should remove most constant uniform background noise.

 

Otherwise if you are happy to pay there are plenty of noise removal plugins at various price points. Watch some YouTube tutorials on them before parting with any cash so you can see and hear how they work so you know which one(s) are going to be the most useful for the type of noise you want to remove.

 

After this is is worth thinking about how you are making the original recordings and take steps to ensure that any new ones are as noise-free as possible. If you aren't recording as much unwanted background noise in the first place removing it will be a lot easier. Also if the you are able to make the phase cancellation method work for you, if in future you find yourself having to record in a very noisy environment, you should spend a few extra minutes recording just the background noise so you have plenty of samples to use for removing it from the speech recordings later on.

[lowdown]

15 hours ago, HornetPinata said:

Ah. Thanks. As you can probably tell I'm lacking in knowledge here.

I'll have a look at Spectral layers and RX 11 Elements, (and maybe even AI though the whole concept freaks me out a bit).

Having online tutorials to should make it easier to grasp things and as there are quite a few recordings to go through 68 pounds might be a good deal if I get my head around how to use it.

I'll have a look on all your suggestions tonight.

 

Thank you again for all your advice and help.

 

I have only just upgraded from SpectraLayers 10 to version 11 myself, so I have been watching videos on what's new when I stumbled across this channel which you might find useful. 

It's about 20 minutes long with the fellow doing some comparisons, and he's also put together some audio cleaning in action, specifically voice related. 

 

 

[HornetPinata]

Thank you @BigRedX and thanks again @lowdown 

[Wolfram]

Steinberg Spectralayers is the tool you want - you don't need a(nother) DAW if it's speech. I'm still on version 10 but 11 has just been released and I will upgrade. I've also used iZotope RX but not for a few versions now; I found Spectralayers' approach far more intuitive and effective for a musician so have stuck with it.

 

You'll probably find it useful for other tasks too. I recently used it to remove the lead vocal from a song to create a practice backing track and pitch that down two tones to bring it into the range of a singer for practice. Job done in a couple of minutes to a very high standard.

[HornetPinata]

Thank you @Wolfram!

I'm sorry for not replying sooner, life just happened to go nuts on me so I've not seen your reply before now.

I'll give Steinberg Spectralayers another look when there is time. 

 

Thanks for the advice and taking the time to answer.

[Dood]

I use Waves Clarity VX and Izotope's RX Essentials. These are the budget options of their bigger plugin suites. I don't think I ever even paid full price for these either - you may as well buy the whole Izotope Essentials Suite that includes a whole bunch of plugins that I use all the time. 

I produce videos professionally and between these plugins they cover pretty much all bases, but, as mentioned above, they can't perform miracles. As the ole addage goes, "SISO"... or, "you can't polish a turd".

 

I'll have to try out Spectral Layers, that sounds really good, though I suspect it will have a lot in commin with RX which is no doubt Steiberg's competitor in the noise-management market!

Edited July 12 by Dood