Normalization of sound files. Normalization of soundness and the maximum allowable level of the audio signal Normalization of sound

I recently called upon a wonderful resource of the independent sound engineer Des McKinney. Tse just krinitsya korisnyh articles on the latest sound recording and vidoma! The first axis is one of them (strong shift by):

The normalization process is often used by newcomers to digital sound recording. As a matter of fact, it seems, "normalization" is of little importance, it’s natural, it’s like a pantelic. However, newcomers and fakhivtsi can be full of myths and misinformation, which are enough on the subject.
I will tell you about 10 wider pardons, and the fact that we will get the truth.

Pikova Normalization

For a cob, actions are explained: So yak "normalization", which can mean a few speeches (Div. Lower), the myth below us should lie before the peak normalization.

Pikova normalization is the process of automating processes, which changes the level of a skin sample in the signals of a digital sound recording in equal measure, such a rank, which is the best sample that reaches the designated level. Zvychay, the process is victorious in order for the signal to reach the maximum of 0dB - the highest level allowed in digital sound.
The process of normalization is similar to the repositioning of the knob of the mucousness of the fader: the entire signal is changed by the same “unruffled” bit, up or down. With normalization, the system will know the most powerful peak and the new vibrant sound.

The deyak from the myths in the lower image does not represent anything else, as the process is not disturbing, as it will be done with wider pardons, albeit not the least, because the myths look like more fundamental sound, the sound is in the sound of digital

Mifi and misinformation.

Myth # 1: For normalization of several tracks, the stench to sound with the same sound.

Normalization of some tracks to the out-of-the-box level is guaranteed only for the sink, but the tracks are identical. However, our sensibility lies in various factors, including the intensity of sound, triviality and frequency. A good signal for the signal is important, but it can’t be directly attributed to the fullness of the track.

Myth # 2: Normalization To rob the track on the floor with a gooey one, it’s worth it.

Listen to ci two mp3 file, skin normalization up to -3dB:

If the level of the track is low, then you can’t even pick up the gain and muck controls, so you can’t beat the track down. In case of a problem with the recording, in the ideal you can rewrite the track with the necessary amount ... If not, normalization can be done in a short take.

If there is a need to set the right track to the track without changing it, it’s harder. For example, it can be done with a vibrating signal, loud noise and non-musical content. Surely, you can set the peak rivin by hand - an audible track and every single one of them ... let the robot for you can develop the normalization function.

Myth # 9: Normalization of the guarantor that the track will NOT be clipped.

A single track, normalized to 0dB, so it won't be overwhelmed. However, if the track will be chipped or filtered, it will appear as a result of the mutilation. The first track is a part of the mix, which includes some of the tracks, normalized to 0dB, and a guarantee that all the tracks must be replaced by the best pick of any one. In other words, the normalization of the capture of you appears to be the rearrangement only in the simplest way.

Myth # 10: Normalization of a dither operation.

Tsei ostan_y myph є slightly esoteric, ale vіn zrіdka appears in the fenestration of the recording. Say, at the form, say: "It’s okay to normalize in 24 beats, but not in 16 beats, so ..."

I’ll just say: dithering will stagnate when the size changes. (For example, revision from 24 bit to 16 bit). Normalization f, pratsyuє directly from the bite glybin, changes only to the level of the skin sample. Oscillations, any changes in the size are not displayed, dithering is not required.

Інші Visnachennya.

Normalization can mean a few of these speeches. In the context of mastering an album, engineering often normalizes tracks to an album of the same level. It’s worth noting to the point of getting rid of the mucousness and not getting to the peak of the skin track.

The system deyak (for example, SoundForge) promotes the "RMS Normalization", which is based on the calculated root-mean-square value. It’s about getting rid of foodiness, and it’s also not going to lie in the pea rivnya. However, both as a result of the normalization, the process of comprehension is also reflected in the way.

Sound normalization in Sound Forge Pro 10

Take a rivin sound signal without risk to waste the quality, allowing the function "Normalize"... Algorithm of її robots in the offensive: the program is virakhovu the level of the signal from the level of the maximum possible signal, the backward level of the file on the basis of the difference. schob hurry up function "Normalize" the dialog is shown in the menu item "Process". The main parameter є "Normalize to", I will give the maximum possible signal to the signal, as if it will be normalization of sound in Sound Forge.

It is very easy to normalize the signals of decal files, which is very clear when recording a CD. For tsyogo, pressing the button "Scan levels", We will scan the audio file, for the soundness of which will be equalized. Then the onset audio file is displayed in the dialogue window "Normalize" "Use current scan level (do not scan selection)"... Onslaught on the "OK" button in the dialogue window "Normalize to"... program to break normalization of food in audio files.

function "Normalize" you can also visonuvati more folding obrobka, counting the middle "I will take away the foodiness"... It’s just a certain situation, if there’s some kind of sound in an audio file, if it’s overwhelmed at one level of gunosty with the smallest, it’s sound like a voice. The reason for this is the power of human hearing. Sound Forge is able to visualize data from a file's point of view. For the whole dialogue window "Normalize to" you need to activate the remix ... At the same time, more parameters will be available: "Ignore below"- the value of the tsy parameter of the value of the threshold of the permissible level for the sound. All values ​​lower than the specified threshold, when scanned "average, take the muck", will be ignored. A large value for this parameter is approximately "-45 Db".

parameter "Attack time" In the case of the program, as soon as possible, the digital gateway is sent to the signal, so that the sound is valid for the scanned data. To that, in the audio file, it is often the case that one sounds, for example, drumsticks, then stand up given value yaknaynizhche, in the last part of the sound, there will be no sound. A large number of vapors have a value of less than 200 milliseconds to cope with the decrees of the directors.

parameter "Release time" vkazu programs, as quickly as possible, close the digital gateway. If, when scanned, it is necessary to pay more money for more material, then set a higher value of the parameter.

As a result of the singing intercourse of human hearing, even more and more low frequencies slightly more important, below average. The position can be corrected by standing up as a great warrior Use equal loudness contour. Qia function because of the inaccurate spectrum in frequencies, so in most cases it is worthwhile to fix it.

To set all the characteristics of the onslaught, just click "Scan levels", Start the scan process "Sleep in guchnost".

With robotic functions "Average RMS level (loudness)", Slide be careful, adjust the parameter "Normalize to", so as a vibration of even a higher value can lead to deformation of the sound, or when the data is displayed. As you can’t change the value of "-6 Db", it will turn on.

For the quickest way to see the message, you can vibrate the item "Apply dynamic compression", Which is in the bookmark "If clipping occurs".

Onslaught is the "OK" button. The program normalizes the soundness of the audio, the value of the current parameters.

(0)

Knowing about Sound Forge Pro 10
	1.	interface	3:13	4	25906
	2.	Navigation in Sound Forge Pro 10	2:00	0	8937
	3.	markers	1:50	0	6369
	4.	oblasts	4:23	0	5093
	5.	Poshuk	4:01	0	4476
Sound Forge Pro 10 editing basics
	6.	Instrument "Magnify"	1:21	0	7006
	7.	seeing	1:41	0	4729
	8.	Copy and paste functions. Part 1.	3:20	0	7352
	9.	Copy and paste functions. Part 2.	3:20	2	44121
	10.	skasuvannya diy	2:45	0	2863
	11.	Іnstrument "Olivets"	3:16	0	5686
functions
	12.	Zsuv on the axis of the amplitude	2:16	0	5631
	13.	change of size	2:17	0	4908
	14.	The frequency of sampling	9:33	0	5605
	15.	Fragments in silence	4:41	0	4850
	16.	insert silence	1:05	0	3643
	17.	Snake to sound. Part 1.	1:09	0	8053
	18.	Snake to sound. Part 2.	1:09	0	9455
	19.	normalization to sound	2:37	0	27550
	20.	change of channels	4:31	0	4636
	21.	panorama	3:26	0	3538
	22.	Equalization, part 1.	2:12	0	5807
	23.	Equalization, part 2.	2:12	0	5087
	24.	Equalization, part 3.	2:12	3	3056
	25.	Turning back	3:20	0	4656
	26.	zmіna shvidkostі	1:57	0	18614
Effectiveness in Sound Forge Pro 10
	27.	efficient exo-camera	2:21	0	5897
	28.	Multi-Tap Delay	3:51	0	3021
	29.	Chorus	2:09	0	3275
	30.	Flange	2:25	0	2491
	31.	Pitch bend	2:42	0	3149
	32.	Pitch Shift	3:08	0	12197
	33.	Vibrato	2:47	0	2171

The normalization of audio signals by peaks led to significant differences in the density between the channels of the change;

Shown as standardized in European regions by document EBU Tech Doc 3205-E

With a fast growth of digital phonogram production and digital distribution of audio content, the maximum permissible audio signal standard is specified by the ITU-R BS. 645, I didn’t see it for some reason, and I experienced it for myself;

ITU-R BS. 1770 the international standard for soundness of audio programs was introduced, new parameter the audio signal is one of the finest things.

As a matter of fact, the European Union recommends that when the audio signal changes, the new unit LU (Loudness Unit) and LUFS (the unit of purity of the current scale) is recommended. (Naming "LUFS", based on the international convention on terminology and equivalently naming LKFS, as per ITU-R BS.1770-2).
It is recommended for the general characteristics of the transmission to carry out the three main parameters:

- Program Loudness;
- Loudness Range;
- Maximum True Peak Level.

The basic rules for setting the parameters before the next points:

For the nominal value of the program, the document EBU R 128 is recommended to accept the level of the level of -23 LUFS, and in case of cases, if the adjustment of the nominal level is unreliable (for example, in case of live translation), the allowance for the change in the nominal value is 1.0.

The audio signal of the transmission is guilty of being, as a rule, viming as one whole without seeing any specific fragments, such as mova, music, sound effects.

The maximum allowable mittєvii transmission rate is guilty of but іvniy -1 dBTP (decibels of true peak).

All accusations are carried out in the form of specific documents: ITU-R BS.1770, EBU Tech Doc 3341 and EBU Tech Doc 3342.

* EBU - European Broadcasting Union

For confirmation, by members of the EBU (EBC) in Russia є only "First Channel", "VGTRK", "Radio Mayak", "Orpheus", "Voice of Russia". Only gadati are set by the same standards.

In addition, the archive file with the EBU documents has a Russian translation, and itself:

EBU Tech 3341;
EBU Tech 3342;
EBU Tech 3343;
EBU Tech 3344;
Esse_625b2- ese Anatoly Sokolin: "Revolution, yak shocked by the audio";
R68_2000_EBU- technical recommendation EBU R68-2000. The installation r_ven in the possession of digital audio and digital audio recorder;
EBU R1771- vimogi before adjusting, as well as being rich and genuine;
EBU R1770-1- recommendation MCE-R BS.1770-1. Algorithms for vimіryuvannya guchnostі sound programsі true peak sound signal;

Here you can always edit the actual original documents.

Everyone is welcome! Normalization of the sound is not a problem in order for Audacity to navigate on the cobblestones.

It's almost a value.

Normal sound - tse, in a simple way, so turn it into an audio editor, you can hear it, but itself:

• clean up the background noise,
• viralize the soundness of the move by stretching out the sound track,
• pick up some great wikidi / piquancy,
• clean up some sounds (cough, for example),
• The soundness of the recording is so loud that you can hear it comfortably on all types of computers and mobile annexes, Vistavlyayuchi guchn_st attachment to the middle rivn.

How important is it? Even more important! Garne video with a nasty sound - pennies for the wind. Video "steer" in internet marketing. Sell ​​your videos through your own online store, send your services through the net, if you will have a corporate website, and use your channel on Youtube - everywhere you need to have a decent video. Ale video, then video, and if you have a sound suprovid will be quiet, deaf, with noises, other defects, then respect, that the whole robot is not. Nichto take video for 10 seconds will not continue to wonder.

I’ll say right away that if you are encouraged to take a professional video camera on your super-happy road, then it’s priceless. Make a noise, then write down more beautifully than a smartphone. So the sound is 100% "vityagnuti" with a pershoklazny "zalizom" not vyde.

Professionals to repent for sound editors... Take okremo audio performance and edit її. At the end of the day, I’m going to tell you how to use Audacity to normalize the sound.

Why Audacity? That is why:

1. Specialized program - audio editor sound files.
2. It’s hard to finish it in order to make it sound like a good sound.
3. Bezkoshtovna.
4. Dosage is easy to learn. Especially if on the right you have to use standard awkward operations with sound.

Well, let's do it pretty well.

You can find out from the statistics:

In order for everything to be as close as possible to the real life and intelligence, video recording, destruction on the very own smartphone - htc one v. Video is known with HD permission. For the current year it is not different, but the standard. The sound will start like a smartphone - if it's close, then it's good, if it's in the country, then it's even in the middle.
So, our nipers task:

Yak vityagti sound from video to an okremiy sound file

Sposobiv - masa. Schob not zharaschuvati post with other row details, I will briefly describe all about three. Viberate for yourself.

1. For help bice paid programs Freemake Video Converter
2. Additional paid programs Total Video Converter
3. For help, you have a video editor. But it’s your fault. Especially if the whole or part of your business is on the Internet. Especially if you regularly know and video on your site. Yaksho vi want vicladati garne video, Shchob yogo marveled at a lot of people.

First two points explain in detail not varto. Everything is absolutely awkward there, but if there will be problems - write, I will explain.

The axis on the video editor is detailed. In sense yak vityagti, the sound is from the video to help. Video editors can be even more abundant. I use one of the most popular ones - Sony Vegas.

Copy the video from your smartphone to your computer.

Vidkrivaєmo video editor.

Via the File menu - Display the video file.

і vibrate format to mp3 file. Onslaught on Custom ...

and vibrate saving parameters. I recommend vibration Mono, bit rate 128 kbps and frequency 44 100 Hz.

I will name the save folder and name it in front of the front mp3 file.

All audio recordings have been saved and are now safe normal sound... I’ll write everything down.

Croc 1. The first time the plug-in Hard Limiter was stored

On the recorded sound, you can see the wikidi pikes on the mundane. If you don’t change it, then it’s great to pull the stench, or you can knock it out. It could be a cough, and a raptor voice put the stiletto in, a signal to order a car and so far. Tom:

Kilkom in the region of managing the authorities of the road with the help of the button of the bear and seeing the whole road by ourselves

Then, in the menu Effekti-Hard Limiter ... and set the axis of such parameters

Kilkoma OK. Ready.

Croc 2. Normalization to sound

Sound recording from microphones, smartphones, dictaphones to go quietly in order to direct the axis in such a view of the video from the view of video on YouTube. So we need to get the sound of the sound. Ale bazhano zrobiti tse so, the sound will be heard, but not as given. For a wide range of functions, the Normalization signal plugin is used. The wine is very strong, albeit so, the maximum amplitude is fixed. For the whole image in the menu Effective-Normalization to the signal ... Set in the end -3.0 db.

Tisnemo OK. Marvel at the result.

Croc 3. Processing the audio file with the Compressor plugin ...

prodovzhuєmo Audacity for normalization of sound and the Compressor plugin is completely mastered ... Why is the Compressor needed ...? The compressor averages out, changing the difference between the quietest and most reliable dilyankas. Buvag, lyudin speaks at the microphone now more loudly, now more quietly, and when there is a great increase in hearing, it’s uncomfortable to record. Stirring with a compressor, the sound of the voice grows bigger, without streaks.

Otzhe ydemo in Efekti-Compressor ...

and embossed OK. Radієmo to the result.

Croc 4. Finish processing with Hard Limiter plug-in ...

Yak bi is good without cutting off the sound of the Compressor, the algorithm may have some shortcomings, and when singing, I know the sound. Once again, use the Hard Limiter plugin ..., only set the level not -10, for the first time, but -2.0 db.

Axis in general and everything. In most cases, 4 crocs are enough. Now more foldable drops are visible, but itself:

1. even in front of the plugins - Normalizing the signal ... and the compressor ...
2. і how did the record go when high rivn background noise - by handing the refrigerator, the fan is noisy, like a good buzz.

Manually virivnyuvannya guchnosti okremikh audio tracks

For a very simple plug-in Potential signal. Win din yak pen of guchnost. At the end of the stage, it is correct, as the sound of the signal through the Normuvannya signal ... and the Compressor ... and in general is a virivnyana without striking diagrams. Tilki, yak bachit won globally on the great dilyanka. With such a "layout" in front of the plug, do not expect to cope well; I will respect that such a situation is infrequent.

Otzhe, seeing that part of the soundtrack, the signal is clearly lower. Ydemo in the menu Effectiveness-Strengthening the signal ... and by choosing the level of strength to achieve the same fragments in the audio recording according to the guchnost. Marvel at the video, how to see.

Yak tidy up the noise iz I will record

Beastly respect, at once I will explain how I fight myself with uninterrupted background noise. As soon as raptom for an hour without recording a middle one, he coughed, chuckled, and fell - all the noise was not background noise, and if you didn’t need to clean it up, then it’s necessary to do it in the most appropriate ways. And at once we will clean up the background noise itself. Then, just clean up the noise from the audio, it is necessary to know on the audio output the silence, see it and respectfully listen. Bazhano, it’s just the same background noise, without clattering and other "vipadayut" or seeing shmatochki. More beautifully we pick up such a fragment, so that the program can be used to clean up all the audio tracks.

For the whole visually vibrate on the diagrams of a dilenka with a zero, or even close to that, amplitude and vision of the target. Kilkoma on the Play button in the Audacity buttons panel is respectfully audible. yaksho na background noise If only the sounds are lonely, then we know and see a fragment without them.

Know the nicest fragment, you see it. Ydemo in the menu Effective-Visual noise-Create model to noise.

Let's see the whole road. Ydemo in the Effective-Visual Noise menu. We have an axis of ci parameters

The only parameter that can be used to experiment is to suppress the noise. Naipersche field. Raju get loose between 12-24 db. If the rate is lower than 12, then the noise can be reduced slightly. As soon as you grow up to 24, then you can appear on the doorstep, the sound of the dilyankas has become too much.
Wondering about the video, de I keep trying:

From and everything. The audio recording is normalized, having lost the preservation of the viewer file.

Saving Audacity of the road in an audio file

Fighting through the File-Export menu ... I wrap up your respect, but through the File-Save project menu ... you save the audio recording in the Audacity and only format. In order to save it in mp3 or wav format, you need to copy it to Eksportuvati ... Everything is simple. Vibrate the type of file. If necessary - click the Parameters ... and set the required parameters. For example, if you are exporting to mp3, then you can vibrate through the Sound Quality Parameters. I recommend not working below 80 kb / s and up to 128 kb / s. Tse for the voice, zychayno. Yaksho vi wrote music and you need maximum yield sounding, you can deliver 320 kb / s. Mayte only on uvaz, so what is the bittrade (the quality of the sound), the bigger size will be the size of the file.

From the same post, yak Audacity in part of normalization of sound.

Updated in breast 2018 for rock - Tsya stattya was written in 2014. Over the past 5 years, until the end of 2018, the accumulated information has accumulated, subtle and priyomi have appeared, such as:

1. simplify the procedure,
2. change the hour for sound morphing
3. brighten the quality of the bag sound

P.S. Would you like to update your blog post about new articles? Press the qiu button -

5 / 5 ( 1 voice)

For quietly, hto, after all, hearing the music in the mp3 player, melodiously know the problem of normalization sound new compositions for adding a folded playlist. Those sounding stavati is zanadto gooey, then navpaki so quiet, so that it is not possible to add more tunicity, not allowing melody to melt.
I have a problem. most hour did not allow listening to music comfortably.

The first time I started to shukati is a way for the development of the normalization of guchnost. Having been quick with a joke on the Internet, I was able to know the simpler solution with mp3gain programs in the distance. It is also obvious that the program can be victorious without any problems.

Surrounding the simplicity of the victorious, the program is also good and not to introduce any changes in the sound of the data. Vona just change the value of the ID3 tags. Lots of programs (such as iTunes) and mp3 players; Otzhe, the essence of the problem is intelligence, the instrument of knowledge. Proceed.

For a cob I will add the mp3gain program (you can download it http://mp3gain.sourceforge.net/download.php) I will install it on the computer. program will be ready різні movi, Including Russian, but also change the language menu and vibrate Russian.

For a selection, you need to vibrate the folder with files for processing, for which you select the button Submit the folder.

If you want the file to be locked, press the button analysis track

It will take you an hour to complete the procedure.
The program depends on the soundness of the files and the amount of sound you need to adjust the strength of the sound in the tracks. In order to analyze the future of completions, it is possible to bring the drunkenness to one meaning simply by saying it in the field norm of guchnost

In order for it to be finished, all the tracks will sound with the same sound. If the effect of normalization is not possible, and if you bring everything to the cob value, then it will not become difficult to grow. You just need to vibrate in the menu, the item - change rіven і natisnuti - Skasuvati Zmіna Rivnya .

From and everything. To be overwhelmed by the method, the descriptions by me in the statistics, help you
viralize the problem of normalizing the purity in the music tracks and not getting used to music until the constant regulation of the purity in the pleura.

Here I just showed the butt yak pratsyuvati z program. Tse does not mean that it is necessary to work this way. For simplicity of demonstration, the boules are displayed as the maximum value. Tse can lead to a meaningful story. Therefore, choose the values ​​that are acceptable for you.

Specially for quick mastering of material on robots with the program I made a video clip - how to make the tracks sound the same: