Verdict on the settings for new Noise Reduction?

[Paul L]

Default smoothing will be 6 in future versions.

I am pleased we finally increased default smoothing, but don't you think it looks odd to have a default at the extreme of a slider scale?

Gale

My thoughts too.
The extreme should in any case be higher - 10 or 12.
If I'm correct, 6 is approx. 150 Hz and we used up to 500 Hz in the old effect. This might not be necessary anymore with the partially new algorithm but who knows.
What about a track with 16 kHz sample rate?

Robert

Remember that frequency smoothing counts bands now. The equivalence of 6 to the old 150 Hz assumes a 44100 sample rate.

[Gale Andrews]

Default smoothing will be 6 in future versions.

I am pleased we finally increased default smoothing, but don't you think it looks odd to have a default at the extreme of a slider scale?

Gale

My thoughts too.
The extreme should in any case be higher - 10 or 12.
If I'm correct, 6 is approx. 150 Hz and we used up to 500 Hz in the old effect. This might not be necessary anymore with the partially new algorithm but who knows.
What about a track with 16 kHz sample rate?

Robert

Remember that frequency smoothing counts bands now. The equivalence of 6 to the old 150 Hz assumes a 44100 sample rate.

So what is the equivalence of 6 at for example 96000 Hz, or 8000 Hz or 16000 Hz (which a voice track might be)?

It seems you are still discouraging frequency smoothing above 6 by setting 6 as the maximum. Are you so discouraging?

https://soundcloud.com/blizzkrut/sets/a ... nd-problem which is voice with heavy noise (or a normalized version with DC offset removed) needs more than smoothing of 6 in my opinion to reduce artifacts on the voice in the noise-reduced result.

What settings would you use on that? The Forum topic those samples comes from is http://forum.audacityteam.org/viewtopic ... 46&t=87385.


Gale

[Robert J. H.]

Default smoothing will be 6 in future versions.

I am pleased we finally increased default smoothing, but don't you think it looks odd to have a default at the extreme of a slider scale?

Gale

My thoughts too.
The extreme should in any case be higher - 10 or 12.
If I'm correct, 6 is approx. 150 Hz and we used up to 500 Hz in the old effect. This might not be necessary anymore with the partially new algorithm but who knows.
What about a track with 16 kHz sample rate?

Robert

Remember that frequency smoothing counts bands now. The equivalence of 6 to the old 150 Hz assumes a 44100 sample rate.

So what is the equivalence of 6 at for example 96000 Hz, or 8000 Hz or 16000 Hz (which a voice track might be)?

It seems you are still discouraging frequency smoothing above 6 by setting 6 as the maximum. Are you so discouraging?

https://soundcloud.com/blizzkrut/sets/a ... nd-problem which is voice with heavy noise (or a normalized version with DC offset removed) needs more than smoothing of 6 in my opinion to reduce artifacts on the voice in the noise-reduced result.

What settings would you use on that? The Forum topic those samples comes from is http://forum.audacityteam.org/viewtopic ... 46&t=87385.


Gale

If I don't err, the smoothing over 7 bands (since factor 0 = 1 band) for different sample rates are as follows - all in Hertz, except last column:

Code: Select all

Sample Rate	Bandwidth	Smoothing factor
8000	27.34375	37.5875
16000	54.6875	18.29375
22050	75.36621094	13
32000	109.375	8.646875
44100	150.7324219	6
48000	164.0625	5.43125
88200	301.4648438	2.5
96000	328.125	2.215625

The last number shows the factor needed to get the same 150 Hz band width as with 44100/6.

Of course, the window length in seconds changes as well and thus the results are presumably not comparable - I haven't made all the tests needed.

Robert

[steve]

It seems you are still discouraging frequency smoothing above 6 by setting 6 as the maximum. Are you so discouraging?

The higher the "smoothing" factor, the more damage is done by the effect.
It may not be so noticeable with speech recordings. but with high quality music recordings the damage is very evident with the new default settings if you listen to the decay of notes.

As a simple demonstration:

1) Generate a few seconds of Pink Noise and Normalize to -46 dB (about -60 dB RMS). This is a reasonable good approximation of low level noise that one might want to remove from a good quality original recording.

2) Import a really good quality music recording that has very low noise - piano music would be a good choice for demonstration purposes.

3) Create a noise sample from the Pink Noise

4a) Apply to the piano music with "Residue" selected. Note how much of the music is present in the residue.
4b) Apply to a mix of the noise and the music - notice how the notes decay much faster than the original.

[Paul L]

It seems you are still discouraging frequency smoothing above 6 by setting 6 as the maximum. Are you so discouraging?

The higher the "smoothing" factor, the more damage is done by the effect.
It may not be so noticeable with speech recordings. but with high quality music recordings the damage is very evident with the new default settings if you listen to the decay of notes.

As a simple demonstration:

1) Generate a few seconds of Pink Noise and Normalize to -46 dB (about -60 dB RMS). This is a reasonable good approximation of low level noise that one might want to remove from a good quality original recording.

2) Import a really good quality music recording that has very low noise - piano music would be a good choice for demonstration purposes.

3) Create a noise sample from the Pink Noise

4a) Apply to the piano music with "Residue" selected. Note how much of the music is present in the residue.
4b) Apply to a mix of the noise and the music - notice how the notes decay much faster than the original.

Ah well, that might be remedied by allowing the attack and release sliders to be visible again, (release more so than attack,) which is another debate.

[steve]

Ah well, that might be remedied by allowing the attack and release sliders to be visible again, (release more so than attack,) which is another debate.

It's a debate that we had about 12 months ago.
A problem adding more controls is the high degree of interdependency between controls.

don't you think it looks odd to have a default at the extreme of a slider scale?

Yes. We have jumped from one extreme of the trade-off to the other.

[waxcylinder]

My 2c worth

I shall certainly be sticking with the 12,6,0 settings - it's been producing excellent results on the Dublin City FM music shows that I record and post online weekly for a couple of DJ friends.

No-one has complained about the quality of the audio.

Personally I don't think we should have changed this default setting - but sadly it's probably too late now for 2.1.2 now that we are in freeze mode.

We may need to change the advice in the Manual to advise users to choke down the setting for music quality.

Peter

[Gale Andrews]

I shall certainly be sticking with the 12,6,0 settings - it's been producing excellent results on the Dublin City FM music shows that I record and post online weekly for a couple of DJ friends.

No-one has complained about the quality of the audio.

Personally I don't think we should have changed this default setting - but sadly it's probably too late now for 2.1.2 now that we are in freeze mode.

You presumably have high quality audio with very little noise.

For people who really need noise removal, frequency smoothing of 0 was a poor default. We've had too many complaints about it.

I thought smoothing of 6 as a default was OK, it was the same smoothing that we had before at 44100 Hz. I think the new effect needs smoothing more than the old effect did - that is definitely so on the material I work with, which is very quiet music where noise must be dealt with.

With so many use cases, we should not have a smoothing default at the end of the scale, and should not have a much lower slider maximum than we had before. A default of 3 with a slider maximum of 12 would be acceptable too, I think. Importantly, that encourages experimentation in both slider directions, and suggests less may be better than more.

We may need to change the advice in the Manual to advise users to choke down the setting for music quality.

The advice may need to be subtly changed, yes. I think the current advice assumes music rather than speech, but clearly a lot of users are working with speech.


Gale

[Gale Andrews]

It seems you are still discouraging frequency smoothing above 6 by setting 6 as the maximum. Are you so discouraging?

The higher the "smoothing" factor, the more damage is done by the effect.
It may not be so noticeable with speech recordings. but with high quality music recordings the damage is very evident with the new default settings if you listen to the decay of notes.

As a simple demonstration:

1) Generate a few seconds of Pink Noise and Normalize to -46 dB (about -60 dB RMS). This is a reasonable good approximation of low level noise that one might want to remove from a good quality original recording.

2) Import a really good quality music recording that has very low noise - piano music would be a good choice for demonstration purposes.

3) Create a noise sample from the Pink Noise

4a) Apply to the piano music with "Residue" selected. Note how much of the music is present in the residue.
4b) Apply to a mix of the noise and the music - notice how the notes decay much faster than the original.

Yes, I know. I dislike that, and I sometimes miss the ability to control the release.

But the whole thing is a trade off. With higher noise levels, or quieter music, the loss of attack can be subjectively less bad than artifacting.

Ah well, that might be remedied by allowing the attack and release sliders to be visible again, (release more so than attack,) which is another debate.
It's a debate that we had about 12 months ago.
A problem adding more controls is the high degree of interdependency between controls.

It is a problem, but more controls could be hidden behind an "Advanced" checkbox, or perhaps separated into another box.

Also do we have to have the interdependency? Could some filter to mask artifacting be applied "over the top" of the effect, like some kind of post-processing?


Gale

[steve]

Also do we have to have the interdependency?

That's an interesting question. It's really a question of design.

We now have a noise reduction algorithm that works better than the old Noise Removal effect, but users find it more difficult to interact with.
The algorithm has many interdependent parameters, but it should really not be a concern to the user what the algorithm is, precisely how it works, or how its parameters interact with each other. The user just needs to be able to use the effect - preferably without needing to read a manual. At one end there is the algorithm with lots of interdependent parameters - at the other end (hopefully) is the user interface in which "A" leads intuitively to "B".