Ways of processing Mp3, Mp4, FLAC, Ogg, APE, AAC and Wav files can be divided into 2 sorts: distorting and not distorting. And if distorting ways of processing change an original relation and level of amplitudes and frequencies of a sound whereas not distorting ways of processing change level of all amplitudes equally or leave invariable level of all amplitudes and frequencies of a sound.
The most widespread distorting way of processing is а equalizer, and the most widespread not distorting way of processing is а normalizer.
We will consider a normalizer in given article.
Gist of a normalization consists that an amplitude of initial beep varies, and a form remains former.
And it will feel on hearing as change of force a sound or volume.
Change of volume is accompanied by change of dynamic range that is relations of the loudest value to the most silent value of beep.
Increase of a volume leads to increase a dynamic range.
Contraction and reduction of a dynamic range is accompanied by music which sounds exactly, monotonously, all time with approximately identical loudness, it becomes boring, inexpressive, more plane, an expression and dynamism disappears, brightness of perception is lost.
Certainly, it is possible to take advantage of a pen or a button of volume and easier to increase volume level, not resorting to a help of a normalizer but if you have many composition it supplies notable inconvenience. Each composition has its a dynamic range and a volume average level.
If all musical compositions are executed in one style or a genre that, as a rule, it is possible will be limited to peak normalization. Peak normalization for group of compositions fixes for each composition a dynamic range at one level thus that all compositions have a same dynamic range.
If all musical compositions are executed in different style or a genre that, as a rule, resort to normalization on an average level. Normalization on an average level for group of compositions fixes a dynamic range at different level for each composition depending on average value of volume of this composition that creates sensation of identical volume at transition from one composition to another for all group of compositions.
Let's consider normalization of a sound with reference to program Sound Normalizer.
Sound Normalizer differs from other sound editors of type "Sound Forge" which, fulfilling normalization of files Mp3, at first convert this Mp3 file in Wav file, then normalize this Wav file, and then already back convert this file Wav in Mp3 file. Sound Normalizer fulfills normalization of files Mp3 without conversions or convertings and thus simply changes a control footing in each frame of Mp3 file.
Sound Normalizer differs from other players of type "foobar 2000" which, fulfilling normalization of files Mp3 use for this purpose APEv2 tags leaving thus each frame of file Mp3 not touched. Sound Normalizer does not use APEv2 tags for normalization that does possible listening Mp3 files in hardware players with flash memory. Besides Sound Normalizer fulfills testing and normalization on each channel separately that allows to align automatically at normalization volume level in each channel on identical level that improve a stereo effect.
You need to open a file prior to a beginning of normalization.
The program has 2 modes: single and batch.
You can open a file in a single mode having executed a command "Open".
You can go in the batch mode having executed a command ”Batch Processor”.
You can open files in the Batch Processor using three ways:
If You execute a command “Add Files” that can select Mp3, Mp4, FLAC, Ogg, APE, AAC, and Wav a file or files;
If You execute a command “Add Folder” that can select all Mp3, Mp4, FLAC, Ogg, APE, AAC, and Wav the files which are in this folder;
If You execute a command “Add Folders” that can select all Mp3, Mp4, FLAC, Ogg, APE, AAC, and Wav the files which are in this folder and all subfolder being in this folder.
The program has for Mp3 files normalization on an average level with definition of clipping. As level of a sound for Mp3 files is near to peak level or exceeds that one peak normalization for Mp3 files it appears not expedient.
Combination of a normalization on an average level with a peak normalization preventing a clipping will be a optimal variant.
Clipping is a distortion of beep expressing in appearance of "hissing" and "cod".
Clipping appears in Mp3 file when Wav file with peak level above 90 % is converting, therefore before converting Wav file in Mp3 carry out peak normalisation Wav file with level no more than 90%.
Please, look and listen the same beep without clipping and with clipping.
No Clipping (the file NoClipping.mp3)
Listen and compare the non-clipped audio "NoClipping.mp3" with the clipped audio "Clipping.mp3"
Clipping (the file Clipping.mp3)
Listen and compare the clipped audio "Clipping.mp3" with the non-clipped audio "NoClipping.mp3"
Before normalization fulfil test for Mp3 files and define a recommended maximum level of normalization without clipping will be the best order of performance of normalization.
The recommended maximum level of normalization without clipping is a optimum level of normalization eliminating clipping and providing maximum quality of a sound.
In the batch mode the recommended maximum level of batch normalization of the list of processing without clipping is the optimum level of normalization eliminating clipping and providing maximum quality of a sound, calculated for a current list of processing.
There is a command ”Normalize each file on a maximum level without clipping” in the batch mode this command represents implementation of peak normalization on a maximum level for Mp3 files.
Level of normalization for Mp3 files expresses in percentage concerning value in 89 db in the program Sound Normalizer. For example 89 db is 100 %.
89 db is defined by practical consideration for majority Mp3 files volume level, on which else there are no clipping.
Reduce Mp3, Wav and FLAC file size
If you need to reduce quality and a size mp3 file then you should use a converter from mp3 in mp3.
Sound Normalizer includes converter from mp3 in mp3 with saving initiating ID3 v1 and v2 tags ( Actor/Title/Genre, etc.).
It is necessary to note that increase of a stream speed or bitrate of a mp3 file does not increase quality of sounding this file.
You will see a column "Estimated size", passing in a batch mode as it is visible in a picture more low, value in this column will be updated each time when you in encoder settings (the button "Settings...", the Output format "MP3" on the tab "Batch Converting") press button "Apply" or "OK".
For reduce size mp3 file is better to reduce stream speed or bitrate. In this case you have 3 variants of a compression mode of the stream data:
1. With constant stream speed (Constant bitrate, CBR);
2. With variable stream speed (Variable bitrate, VBR);
3. With average stream speed (Average bitrate, ABR).
CBR mode will give you predicted size mp3 file.
VBR mode will give you unpredictable a value of reduction mp3 file.
ABR mode will give you predicted with much bigger (in comparison with VBR) accuracy a value of reduction mp3 file.
If you need to save quality and to reduce a size of Wav file then you is better to use the converter from Wav in FLAC. By default encoder FLAC parameters are adjusted on obtaining of an optimal ratio speed/level of compression for a majority of types input files. However this majority of types input files is that that the majority of parameters encoder FLAC are calculated for a sound from CD-audio (i.e. 44.1кГц, 2 channels, 16 bits). Thanks to of simple algorithms of decoder a speed of decoding FLAC above speed of coding besides encoder FLAC has hardware support that does the codec even faster and power efficient as it not strongly loads the central processor. After you converted a sound in a file of format FLAC you by means of program Sound Normalizer can reduce file size .flac still at least by 10 %. For this purpose you need to open a file .flac and to reduce level of normalization of this file proceeding from that that reduction of level of normalization by 1 % reduces file size .flac at least by 0.2 %.
The normalization is not distorting sort of processing.
The normalization is accompanied by change of a volume or of a dynamic range, that is the relation of the loudest value to the most silent value of beep varies.
The normalization allows to improve quality: if it for Mp4, FLAC, Ogg, APE, AAC, Wav files expresses by increase of a dynamic range for Mp3 files it expresses by elimination of a clipping.
Starting normalization Mp3, Mp4, FLAC, Ogg, APE, AAC and Wav files it is necessary to mean, that there are 2 sorts of normalization:
Normalization on an average level.
It is necessary to know that there are 2 modes of normalization:
The normalization of the batch mode works with a list of processing and allows to align perceived volume for group of compositions. If all songs belong to the same genre or style of music in group of compositions that usually enough and peak normalization.
If there are songs in group of compositions having different genres then it is usually necessary to use normalization on an average level.
Usage of normalization on an average level is necessary for Mp3 files as their level of volume, as a rule, is equal maximum or exceeds it, that calls appearance clipping. Mp3 files have a small size and consequently are allocated on a disk of usual capacity in a considerable quantity and belong not seldom to different genres of music. Usage of normalization on an average level together with peak normalization preventing clipping of beep and allowing to receive maximum quality therefore will be optimal for Mp3 files.