Volume regulation and maximum permissible audio signal level. Normalization of sound files Articles about sound normalization and volume

Requirements for television companies to normalize loudness have been put forward for a long time, and after the adoption of new amendments to the Federal Law “On Advertising”, the task of normalizing loudness has become especially relevant.

The amendments introduce requirements for the ratio of the volume level of advertising and the average volume level of a television or radio program interrupted by advertising. Deviation from the required standards threatens television companies with significant fines.

To date, two independent software methods volume normalization and added means of monitoring compliance of the volume level with regulatory requirements.

Real-time volume normalization

Specially developed software plugin "APTO audio processing" designed for automatic normalization of the output audio signal "on the fly" (in real time, without pre-processing) during broadcasting. Below is the plugin settings dialog "APTO audio processing".

It is an additional paid software option.

Main characteristics:

  • Linear Acoustic's adaptive APTO algorithms are used;
  • to comply with the new legal requirements there is a special regime (EBU_R128);
  • minimum set of settings:
    • “target level” – the required value of the output volume level in LUFS;
    • “aggressiveness” – the speed of reaction to changes in sound level: the higher the value, the faster the adjustment to the target level is performed (but also the stronger the sound distortion);
    • “analysis time” – time interval for continuous analysis of sound parameters statistics.
  • Various types signal: analog (CVBS, YUV, Y/C, RGB), SDI, programs in the MPEG-TS transport stream.

Pre-processing audio in files

Program used SLAudioNormalizer from the standard software set. The settings indicate the set of folders in which files should be processed and the mode. Program in automatic mode performs analysis audio track video files. The sound in the file itself does not change, but an auxiliary metadata file is generated with information about the required change in sound volume. Below is the dialog for setting the volume level normalization of the program SLAudioNormalizer.

Main features of the program:

  • several loudness normalization modes, one of which meets current regulatory requirements;
  • start file processing at startup operating system;
  • tracking the appearance of unprocessed files;
  • normalization of multilingual audio;
  • logging program operation;
  • processing files in subfolders;
  • flexible settings for file processing mode.

Volume Controls

The standard set of software for our products includes the program SLLoudnessMMeter– software loudness meter in EBU mode (ITU-R BS-1771 and EBU Tech3341).

Main characteristics:

  • several indicators and time scales: “instant”, “short-term”, “integral”;
  • scale displaying the maximum permissible level of true peaks;
  • various ways to display indicators;
  • selecting the scale display method;
  • selecting the integrated volume measurement mode;
  • the ability to manually reset statistics;
  • recording in the measurement protocol file.

Functions implemented in programs SLAudioNormalizer And SLLoudnessMMeter, are also built into the editor program for viewing video files and setting playback parameters. By opening a video file in this program, you can perform all the necessary volume measurements and select a control value for the sound from this file. Included in the standard software set

Below is a volume measurement block in accordance with the new legal requirements in the SLTrimEditor program

To ensure compliance with the standards set by regulatory authorities for sound volume in television programs, we recommend that our clients use all of the listed tools for sound normalization and control measurements in combination:

  1. Use the SLLoudnessMMeter program at different stages of television production. This will allow the broadcasting company to carry out control measurements in full compliance with the measurement methodology proposed by the Federal Antimonopoly Service of the Russian Federation.
  2. Based on the results of measuring the integral volume of video files, bring the sound volume in the files to the required level using the program SLAudioNormalizer.
  3. Enable constant leveling of the volume of TV shows at the server output using plugin tools "APTO audio processing".

Normalization only using the program SLAudioNormalizer gives the required result in the case of relatively short and “uniform in sound” commercials. If the integrated volume of each commercial is equal to the target level, then the integrated volume of all commercials will be close to this level. But in the case of programs or films, this tool is not enough, since the volume of different fragments of a program or film can vary greatly (the same is true when broadcasting local programs, for example, news, immediately followed by an advertising block). And this is where plugin help is needed. "APTO audio processing", which performs sound processing in real time, “on the fly” reacting to changes in sound level and taking into account accumulated measurement statistics.

Sharing The proposed normalization methods lead to a high-quality result - automatic volume leveling with the least sound distortion throughout the entire broadcast time and, most importantly, when inserting advertising blocks into programs and films.

Normalizing audio signals to peaks resulted in significant differences in loudness between broadcast channels;

The readings of the QPPM, standardized in European countries by EBU Tech Doc 3205-E and commonly used quasi-peak level meter, do not reflect the signal volume, because this device was not originally intended to record the average signal value;

With the rapid growth of digital production of phonograms and digital distribution of audio content, the standardization of the permitted maximum audio signal level, defined by the ITU-R BS document. 645, does not meet modern requirements and has become obsolete;

Document ITU-R BS. 1770 determined international standard measuring the volume of audio programs, introducing a new parameter of the audio signal - the volume unit.

In accordance with the above, the European Broadcasting Union recommends using the new LU (Loudness Unit) and LUFS (loudness unit relative to full scale) when measuring audio signals. (The name “LUFS” corresponds to the international convention on terminology and is equivalent to the name LKFS, which is used by ITU-R BS.1770-2).
Recommended for full characteristics transmissions to measure three main parameters:

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

The basic rules for measuring these parameters boil down to the following points:

EBU R 128 recommends taking a level equal to -23 LUFS as the nominal value of the program loudness, and in cases where precise maintenance of the nominal level is unattainable (for example, during a live broadcast), the permissible deviation from the nominal level should not exceed ± 1.0 LU.

The audio transmission signal should generally be measured as a whole without isolating specific portions such as speech, music or sound effects.

The maximum permissible instantaneous transmit level shall be -1 dBTP (decibels true peak).

All measurements must be made with meters specified in the relevant documents: ITU-R BS.1770, EBU Tech Doc 3341 and EBU Tech Doc 3342.

*EBU - European Broadcasting Union (European Broadcasting Union)

For reference, members of the EBU (EBC) in Russia are only Channel One, VGTRK, Radio Mayak, Orpheus, and Voice of Russia. What standards other broadcasters use is anyone's guess.

Attached is an archive with EBU documents in Russian, namely:

EBU Tech 3341;
EBU Tech 3342;
EBU Tech 3343;
EBU Tech 3344;
Essay_625v2- essay by Anatoly Sokolin: “The revolution that shook the world of audio”;
R68_2000_EBU- EBU technical recommendation R68-2000. Installation level in production equipment digital audio and digital audio recorders;
EBU R1771- requirements for instruments measuring loudness and true peak level;
EBU R1770-1- Recommendation ITU-R BS.1770-1. Loudness measurement algorithms sound programs and true peak audio signal level;

Here you can always get up-to-date original documents.

Hi all! Sound normalization is not a problem for those who know how to use Audacity even at the most basic level.

Let's start with a definition.

To normalize sound is, simply put, to process it in an audio editor so that it is pleasant to listen to, namely:

  • remove background noise,
  • equalize the volume of speech throughout the entire audio track,
  • remove sharp emissions/peaks in volume,
  • remove unwanted sounds (cough, for example),
  • make the recording volume such that it can be comfortably listened to on all types of computers and mobile devices, setting the device volume to medium level.

How important is it? Very important! A good video with bad sound is a waste of money. Video “rules” in Internet marketing. Whether you are selling through your online store, promoting your services online, building a corporate website, or trying to boost your Youtube channel - you need to be able to make a decent video everywhere. But video is video, and if your soundtrack is quiet, muffled, with noise and other defects, then consider all the work in vain. No one will continue to watch such a video for more than 10 seconds.

I’ll say right away that if you rely on your ultra-modern expensive professional video camera, then it’s in vain. It will record noise even better than a smartphone. So you won’t be able to “pull out” the sound 100% with first-class hardware.

Professionals use sound editors for this. They take a separate audio track and edit it. In this post, I will teach you how to use Audacity to normalize audio.

Why Audacity? Because it:

  1. A specialized program is an audio editor for sound files.
  2. Powerful enough to do anything with sound.
  3. Free.
  4. Quite easy to learn. Especially when it comes to standard, simple operations with sound.

Well, let's get started.

From this article you will learn:


In order for everything to be as close as possible to real life and it’s clear, let’s take a video recorded on the most ordinary smartphone - htc one v. He shoots video in HD resolution. Today this is no longer something prohibitive, but a standard. The sound captures like a smartphone - if close, then it’s good, if at a distance, then it’s mediocre.
So, our very first task:

How to extract audio from a video into a separate audio file

There are a lot of ways. In order not to clutter the post with minor details, I’ll briefly tell you about just three. Choose one that is convenient for you.

  1. Through demon paid program Freemake Video Converter
  2. Through the paid program Total Video Converter
  3. Using your existing video editor. And you must have it. Especially if all or part of your business is online. Especially if you regularly shoot and post videos on your website. Of course, if you want to post a good video so that many people will watch it.

The first two points are not worth explaining in detail. Everything there is completely simple, but if you have any problems, write to me and I’ll explain.

Here I will dwell on the video editor in more detail. I mean, how to extract sound from a video using it. There are also a lot of video editors. I use one of the most popular - Sony Vegas.

We copy the captured video from the smartphone to the computer.

Open the video editor.

Using the File – Open menu, open the video file.

and select the format of the saved file mp3. Click on Custom...

and select saving options. I recommend choosing Mono, bitrate 128 kbps and frequency 44,100 Hz.

Select the save folder and the desired name of the saved mp3 file.

We saved the entire audio track separately and now let's start normalize sound. I'll describe everything step by step.

Step 1. Initial use of the Hard Limiter plugin

The recorded sound may contain loudness peaks. If they are not reduced, they can become very annoying or even deafening. It could be a cough, a suddenly loudly moved chair, a horn from a passing car, and so on. That's why:

Click on the track properties control area with the left mouse button and thereby select the entire track

Then go to the Effects-Hard Limiter menu... and set these parameters

Click OK. Ready.

Step 2. Normalize the sound

Usually, recordings from microphones, smartphones, and voice recorders turn out to be quiet so that they can be posted directly in this form as a video on YouTube. So we need to raise the sound volume. But it is advisable to do this so that the sound rises, but not above a given limit. The Signal Normalization plugin is used for this. It increases the volume, but in such a way that the maximum amplitude is fixed. To do this, go to the Effects menu - Signal Normalization... Set the box to -3.0 db.

Click OK. Let's look at the result.

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

Let's continue use Audacity For sound normalization and at this step we will master the Compressor plugin... Please note that you need to process the track in exactly this order step by step, without confusing or skipping. What is a Compressor for...? The compressor averages out, reducing the difference between the quietest and loudest sections. It happens that a person speaks into the microphone either louder or quieter, and if the difference is too big, listening to such a recording is uncomfortable. After processing with a compressor, the voice volume becomes more even, without jumps.

So let's go to Effects-Compressor... Set the same parameters

and click OK. We are happy with the result.

Step 4. Finishing with Hard Limiter plugin...

No matter how well the Compressor processes sound, its algorithm also has shortcomings and under certain conditions it again highlights peaks. To avoid this, process the track again with the Hard Limiter... plugin, just set the level not to -10, as the first time, but to -2.0 db.

That's all. In most cases, these 4 steps are enough. Now let's look at more complex cases, namely:

  1. if the previous plugins - Signal Normalization... and Compressor... - did not do a very good job of normalizing the volume along the entire length of the audio track
  2. and if the recording took place at high level background noise - a refrigerator was working nearby, a fan was making noise, some kind of humming, etc.

Manually leveling the volume of individual sections of an audio track

To do this, use a simple Signal Boost plugin. It acts as a volume knob. At this stage, its use is justified, since the sound has already been driven through Signal Normalization... and Compressor... and in general represents a diagram leveled without jumps. Only, as you can see, it is globally different in large areas. Previous plugins do not always cope well with this “situation”, so now we’ll fix it manually. I note that this situation does not happen often.

So, we highlight that part of the audio track where the signal level is clearly lower. We go to the menu Effects-Signal Gain... and by selecting the gain level we achieve equalization of the audio track fragments in volume. Watch the video to see how this happens.

How to remove noise from a recording

Please note that now I will explain how to deal specifically with continuous background noise. If suddenly during the actual recording someone coughed, sneezed, or something fell, this is not background noise and if you want to remove it, then you need to do it in other ways. Now we will remove the background noise. So, to remove noise from audio, you need to find a section of silence on the audio track, highlight it and listen carefully. It is desirable that it contains only smooth background noise, without clicks or other “dropping out” or prominent pieces. The better we select such a fragment, the better the program will cope with cleaning the entire audio track.

To do this, visually select an area on the diagram with zero or so amplitude and select it with the mouse. Click on the Play button in the Audacity button bar and listen carefully. If there are other isolated sounds in the background noise, then we try to find and highlight a fragment without them.

Having found the best fragment, we select it. Go to the menu Effects-Noise Removal-Create Noise Model.

Then select the entire track. Go to the menu Effects-Noise Removal. We leave these parameters here

The only setting you can experiment with is Noise Reduction. The very first field. I advise you to stay within 12-24 db. If you set it below 12, the noise may decrease quite slightly. If you make it higher than 24, distortion may appear in the remaining areas with sound.
Let's watch the video where I do all this:


That's all. The audio track has been normalized, all that remains is to save it as a file.

Saving an Audacity-processed track as a separate audio file

This is done through the File-Export menu... Please note that through the File-Save Project menu... you will save the audio recording in the Audacity format and only. In order to save in mp3 or wav format you need to use Export... Then everything is simple. Select the desired file type. If necessary, click Settings... and set required parameters. For example, if you export to mp3, then you can select the audio quality through Options. I recommend not making it lower than 80 kbit/s and higher than 128 kbit/s. This is for the voice, of course. If you wrote music and you need maximum sound quality, you can even set it to 320 kbps. Just keep in mind that the higher the bitrate (this is the sound quality), the larger the final file will be.

So, from this post you learned how use Audacity in terms of sound normalization.

Updated December 2018 — This article was written in 2014. Over the past 5 years, by the end of 2018, experience has accumulated, subtleties and techniques have emerged that:

  1. simplify the procedure,
  2. reduces sound processing time and
  3. significantly improves the quality of the final sound
P.S. Would you like to be notified of new articles on this blog? Click on this button -

The computer program allows you to equalize the volume of MP3 music files. The first version of this utility has existed since 2002. The good thing about the program is that it does not require file recoding at all - this allows you to maintain the original sound quality. MP3Gain equalizes the volume level of both a single file and an entire group of files (batch conversion).
We won’t delve too deeply into all the intricacies of the program’s settings and capabilities - we’ll just learn how to easily normalize the volume level in mp3 files without any extra hassles.
We find and .

All the advantages of the program
The program is completely free.
Installs on any version of Windows OS.
Can be used and operated in command line and graphical shell for Windows.
Possibility of batch analysis and file processing.
Normalization occurs without recoding files.
You can convert the same mp3 file many times without the risk of ruining it.
There is a mode for applying normalization only to tracks selected in the working window.
The program fully preserves ID3 tags and file creation dates.
Multilingual interface, including Russian localization.
Localized reference guide on the official site.

Installing MP3Gain
We take the program from SourceForge as an installer. The installation is extremely simple, the only important point is that you need to enable the “Language Files” checkbox, this will install all language localizations of the program, including Russian. If you select “Custom” installation, you can independently select the program’s parking directory.

Setting up MP3Gain
After installing the program, launch it and first of all select the Russian localization of MP3Gain. Next, open the experimental mp3 files. In the program settings, look for the very important item “Change level without clipping” and check it. For brevity, "clipping" is the excess of the signal level, which cuts off the level and re-encodes mp3 files, but we don't need that. And you should also consider the issue of adjusting the volume level. By default, the “Normal” volume is set to 89 decibels (it is better not to change this figure). According to experts, 89.0 dB provides the highest quality results in terms of normalization and elimination of clipping. The rest of the settings are very clear and can be set according to personal preferences or just do everything as shown in the picture. These settings are quite sufficient for simple normalization of the volume level in mp3 files.

Advice! Just in case, you need to make copies of the audio files. MP3Gain does not have the function of saving processed files under a different name; the program overwrites the original ones.

Using MP3Gain
To understand what to do with the two working buttons “Analysis” and “Type”, you need to briefly understand their available modes.
Let's look at the “Track”, “Album” and “Constant” modes.
Track- the program calculates the volume level individually for each track. It then adjusts the volume of each track to the desired level. For example, there are three songs with volume levels of 87, 95 and 91 dB. When using Track Type to bring them down to the required 89 dB level, all of these songs will output at around 89 dB.
Album- the overall volume of the album will be adjusted to the desired level, but the volume differences between tracks in the album will be preserved. For example, there are three songs with volume levels of 87, 91 and 89 dB, the total volume of this album will be about 89 dB. When you apply "Album Type" to bring them to the required level of 92 dB, the program will increase the volume of each of these songs by 3 dB.
Constant- This mode is similar to Album mode. With it, the volume of all tracks simply increases or decreases by a given number of decibels without any normalization relative to each other.

So, let’s conduct an experiment on previously opened mp3 files in the “Track” mode. First of all, we launch file analysis using the “Analysis Track” button. Let's look at the result of the analysis of the source files. In the picture below you can see that in the files “3.mp3” and “5.mp3” there is an excess of the volume level, in other words, there is “clipping”, the letter “Y” appeared in these columns and all the lines turned red. On the contrary, in the file “6.mp3” you can see that it has a reduced volume level.
And then, based on the results of the analysis, in the second action we correct (normalize) this difference in levels by clicking on the “Track Type” button. After successful normalization, which took some time (it all depends on the power of the computer), we look at the resulting result. The last picture shows that the level of all processed mp3 files is very close to the specified value of 89 dB. Those. tracks “3.mp3” and “5.mp3” lowered their volume level, and track “6.mp3”, on the contrary, increased it. That's what needed to be done!