Just as engineers are in no hurry to reveal the secrets of their designs, musicians rarely share anything other than songs with the public. However, sometimes you need not the final material, but the initial one - for example, a vocal part from some composition.
It is necessary
Adobe Audition 3.0
Instructions
Step 1
Install Adobe Audition 3.0. You will also need the Center Channel Extractor VST plugin, which you can download from audio editing sites.
Step 2
Create a copy of the song you want to extract vocals from. It is preferable if it is a composition recorded in high quality format with minimal compression. Open it with Audition.
Step 3
Open the menu Effect -> Stereo Image -> Center Channel Extractor. A window with a dozen sliders for editing the plug-in will appear on the screen. Any song has a unique set of frequencies included in it, and therefore there is no universal meaning for each parameter. To achieve the best quality of the cut acapella, you can only go through several settings.
Step 4
In the Extract Audio From item, specify the time zone in which the vocals are located. The "left" part is the beginning, the "right", respectively, the middle. If you need a specific fragment, then select the "selectively" parameter and set the editing areas.
Step 5
Describe the voice in Frequency Range. The human voice can only sound in a certain frequency range, which you need to define. The user is provided with Male Voice, Female Voice, Bass and Full Spectrum options corresponding to the male, female and bass part, respectively. If there are several vocalists, then it is preferable to choose the last item, which cuts out all frequencies available to human bundles. However, this setting is "coarse" and reduces the final quality.
Step 6
Move Center Channel Level to the right side of the panel.
Step 7
Use Volume Boost Mode at your own discretion.
Step 8
In the Descrimination Settings sub-item there is a package of settings for sound editing. Crossover sets the music level (shift to 0-7%); Phase Discrimination is medium and high; you will have to experiment with Amplitude Discrimination / Bandwidth on a case-by-case basis. Spectral Decay Rate is an integrating and smoothing parameter, its value ranges from 80 to 100%. Select FFT Size in the range 4096 - 10240; Overlays - 3-9; Interval Size - 10-50 ms; Window Width - 30-100%.
Step 9
After changing the settings, click "Edit" and save the mix, from which only the voice remains. If the vocal quality is unsatisfactory, change the settings within the specified range.
