@browj2

That's quite impressive John. Will have to wait and see what the eventual price will be. I would like to see that work in an urban environment.

@Scenestealer

Hi Peter

This company does a wide range of audio separation plugins including a vocal isolator.

Look at the demo videos and try a trial version.

I was quite impressed but they can be quite difficult to tune out the individual bits you need.

@Scenestealer

Hi Peter,

Here is a link for the pricing.

It should work with VPX12 as it comes with VST2 and VPX15 doesn't do VST3. It's free so try it out.

Listen to my video. In the third part, I reduced just the ambient noise and the voice sounds clear. I can get about the same effect using the NR built into MS/VPX using a good noise print, but this tool doesn't need a noise print. I will have to compare them. I just installed Goyo today, so I haven't played much with it. It was recommended on the Vegas forum.

As for others, currently my favourite is iZotope RX10 Repair Assistant that comes with SF Audio Cleaning Lab 4. There are separate tools, like Voice De-noise, but Repair Assistant is a combined one that analyzes the material and proposes the settings that can be adjusted and seems to do it very well. Unfortunately, being VST3 only, they don't work with VPX.

Also with SFACL is SpectraLayers Elements 9 which has noise reduction along with many other things for cleaning audio. It seems very good but Magix has messed up the interface with SFACL as the playback in SFACL does not move the playback marker along properly in SpectraLayers. I have complained about this and several other problems with SFACL to no avail.

Here is a tutorial on SpectraLayers NR:

John CB

@browj2

Hi John

As Ray commented - that is impressive, I will try it on some of my videos where wind noise is a pain for normal reduction techniques.

@Scenestealer

Hi Peter

. . . . typical scenario is recording from a moving mini bus while talking to my camera . . . .

Do you get funny looks from others on the bus. I get funny looks when using the voice control on my DJI Action 3 😦

John EB

As for others, currently my favourite is iZotope RX10 Repair Assistant that comes with SF Audio Cleaning Lab 4. There are separate tools, like Voice De-noise, but Repair Assistant is a combined one that analyzes the material and proposes the settings that can be adjusted and seems to do it very well. Unfortunately, being VST3 only, they don't work with VPX.

 

I am sure you may well know this, but for anyone that doesn't, you can run VST3s in VPX if you use another VST inbetween. So for instance, I use Freestyle from New Sonic Arts, which can load VST2s or VST3s... So, if you open the VST2 of Freestyle in VPX, you can load up any VST3s into it, meaning you will then be running VST3s in VPX....

Izotope's VST 3 plugins can all run in VPX that way. Though again, you may well already know, and it is the additonal step that would put people off wanting to do it that way!

And good tip on the GOYO plugin, I had seen that shown off in a YouTube clip before, was impressed but forgot what it was called! Will probably follow up now, can remember there being some fuss about the always-online situation or data collection etc, so I hesitated at the time and moved on. Seems sensible to download it now to secure the reduced price when the full version is released.

Thanks All for the extra info.

@browj2

Hi John

Yes, that is a good demo in the 3rd part where you have switched it around.

Can you apply it to individual clips or does it only work at track level in the mixer?

Coincidentally I had just installed a copy of SFACL3 to see how the Spectral thing works, but have not delved into it yet. I presume you can open a clip from within MS or VPX and exchange the audio for the modified audio clip automatically.....or maybe not?

I guess I am looking for something that is simple to use on a lot of clips all with varying levels of voice be it me or my friends, and different relative back ground levels. I have tried to do this with the VPX audio cleaning tools and have created and saved Audio effects filters ie." TempoTraveller.afx" using the noise reduction and equaliser settings but just when you think you have made one that works well, you find that is needs more tweaking for the next clip, which is a little cumbersome.

Sadly, I burn myself out tweaking the picture to perfection then find I just do not have the energy (or skills) to improve the dubious audio....😟

@johnebaker

Hi John

"Do you get funny looks from others on the bus." No, because I only travel with funny people (my friends)!

I don't know how you find your Sony FDR AX53, but I find it does not pick up my voice over ambient noise nearly as well as my 2 previous HD cameras (Canon DC10 and Pana SD900) despite my fiddling with the available audio settings - Curiously it has one for "cancelling the shooters voice" and I can't for the life of me figure out when you would want to use it.....wouldn't you just shut your mouth?

Peter

@Scenestealer

Hi Peter,

As they say, people will more easily/readily watch a poor video with great audio than a great video with poor audio.

How long should an image be shown on screen with no audio before one gets tired of seeing it. I use 3 seconds. In one video, I have a somewhat poor image of a couple sitting together on-screen for 30 seconds. The audio is the gentleman describing how they met, then the lady's sister (the lady was deceased) describing what her sister said about how they met. So far, the reaction of viewers is that they keep their eyes on the screen with the static image while listening to the story...and laughing. Audio is important.

Can you apply it to individual clips or does it only work at track level in the mixer?

Unfortunately, MMS/VPX does not yet have VST capability on objects, only at the track level. I tried using the FX Sends channels/tracks in the Mixer and the AUX1/2 sends for objects in the Effects, Audio Effects, General, but was not very successful. So, the best way is to either use separate tracks or send the audio out to SFACL by object or track. That said, Goyo seems to take care of changing audio characteristics as the only parameter that can be changed is the amount of the effect.

I forgot to mention that using the VPX/MMS built-in Noise Reducer, I found that the maximum amount of noise reduction without the audio sounding tinny or with artefacts (Br spelling), was about -8dB. In my Goyo example, I used -10dB and the voice sounded fine.

Coincidentally I had just installed a copy of SFACL3 to see how the Spectral thing works, but have not delved into it yet.

SpectraLayers does not come with SFACL3, only 4. I also have SFACL3 and SpectraLayers shows up there but it works equally poorly (bug reported to Magix). However, SpectraLayers Elements can be opened as a stand-alone and works fine, but I want it to work with SFACL.

However, if you have any sudden noises (clicks, door closing, crashes, etc.) that you want to remove, the Spectral Editor in SFACL is quite good for that.

I presume you can open a clip from within MS or VPX and exchange the audio for the modified audio clip automatically.....or maybe not?

Yes. Replace Music Editor in the Project Settings Folders with SFACL. Follow my instructions in this tutorial:

However, SFACL4 is broken and you cannot use it as the external audio editor in VPX (bug with Magix and nothing done - more users need to complain).

You have to decide what you want to send out - all audio, just 1 or more tracks, or 1 or more contiguous audio objects (in a range and/or limited using Solo).

You may want to cut up the narration parts, add NR to just them to reduce background noise similar to ducking or damping, and no NR on the background. You would need to make audio transitions so that it's not so obvious.

I guess I am looking for something that is simple to use on a lot of clips all with varying levels of voice be it me or my friends, and different relative back ground levels.

Try Goyo on a track like I showed to see if it helps.

However, that is not going to help with varying voice levels. I go through the audio looking for spikes, too loud areas, and adjust the object volume curve, sometimes using Draw mouse mode as it's quicker than manually adding points. Just remember to go back to normal mouse mode.

I also use YouLean Loudness Meter to help locate spikes and level up the audio throughout the video.

Then, of course, there is compression and EQ (low pass filter, de-esser) to go along with Noise Reduction.

Sadly, I burn myself out tweaking the picture to perfection then find I just do not have the energy (or skills) to improve the dubious audio.

I sympathise totally. See my first comment. This is a problem that most of us suffer from, including me, but it's best to spend time improving the audio. Your audience, even if it's just you, will appreciate it.

I find it does not pick up my voice over ambient noise nearly as well as my 2 previous HD cameras (Canon DC10 and Pana SD900) despite my fiddling with the available audio settings -

Try using a lapel mic or wireless lapel mic. This way you can pass the mic around to whoever is going to do some talking. I should follow my own advice.

Curiously it has one for "cancelling the shooters voice" and I can't for the life of me figure out when you would want to use it.....wouldn't you just shut your mouth?

Good one! It seems that some people can't. I have some old VHS footage shot by a friend on a trip to Europe with great narration, "I'm zooming in....now I'm zooming out..."

John CB

@Scenestealer, @browj2, @CubeAce

HI Peter, John

. . . how you find your Sony FDR AX53, but I find it does not pick up my voice over ambient noise nearly as well as my 2 previous HD cameras . . . .

. . . . I can't for the life of me figure out when you would want to use it.....wouldn't you just shut your mouth? . . . .

Mine certainly picked up heavy breathing when held lower than face level when the 'cancelling the shooters voice' option is off.

John - the microphones on this camera are on the top and are surround sound.

The issues I was having with wind (😱) were fixed when I got the Sony ECM-XYST1M stereo microphone, the dead cat on it is very good.

I need a fix now for my DJI Action 3 - Ray has already sent me a link to a potential supplier, since the US supplier of custom dead cat wind shields, I previously used, is no longer in business.

John EB

Hi,

Just a quick update. I opened Goyo in SFACL4 and applied it to a voice recording of me, with -8dB reduction on Ambience. I could detect slight background noise or hiss through my speakers in the original but the audio sounded fine. With Goyo, my voice sounded slighty richer and the background hiss was almost eliminated. Interesting.

@Scenestealer

Hi Peter,

I just noticed that the levels in Goyo can also be positive, so you can increase the ambient sound, keeping the Voice at the same level, or decrease ambient sound, increase Voice and the voice gets louder and stands out better from the background.

John CB

Update from Goyo.

They intend to call the plugin Supertone Clear.

GOYO Beta sign-ups will close on October 1st, 2023. No additional emails will be accepted afterwards.

GOYO Beta will close on November 1st, 2023. You will not be able to use the GOYO Beta plug-in afterwards.

John CB

@browj2 @Scenestealer @johnebaker @Eaglezone

I have found this completely free one. It's also cross platform.

https://ultimatevocalremover.com/

Best to watch some tutorials on it's use.

Ray.

@CubeAce @browj2 @johnebaker @browj2 @Scenestealer

I have the software that Ray lists. It is pretty impressive, once you sort out exactly how to use it.

I confess I have not so far used it to reduce or remove ambient noise from a video but rather to extract stems from recorded audio, e.g., to reduce the level of a lead vocal in order to create a sort of "karaoke/accompaniment" track, or to isolate the drum track or other part of the original "mix".

My personal choice for attempting to isolate "wanted" from "unwanted" audio in a video is iZotope RX10 Advanced, but that is a whole-different ball-game, and price!

Jeff

@emmrecs

Hi Jeff.

What I like is it is constantly being worked on and such software wasn't even possible a few years back. Who knows how good it will get in the next couple of years?

Ray.

@CubeAce

Hi Ray,

Fully agree!

Jeff