A Scientific Approach

Electronic Voice Phenomena UK

Cheshire & Staffordshire ENGLAND

SOUND SAMPLE ANALYSES & FREQUENCY EXTRACTION - ADVANCED USERS.
The intention with this Chapter is to give you the knowledge or a knowledge base to extract certain frequencies for further analyses. If you have read Chapter 1 then you will have heard and seen the make-up of sound. Although I have placed this Chapter in the 'Advanced Users' category, it certainly is within the realms of anyone to quickly learn frequency extraction.

There are a high number of recordings of potential EVP where the investigator is trying, often with the help of friends, to recognise a word or words spoken. If I was to ask you to tell me the colour of the car that's been buried in mud for twenty plus years what would be the first thing you would do? Of course, clean it up a bit so you can recognise the colour. Sound is no different.
Remember my reference to Karaoke Machines in a previous Chapter?. I am now going to quickly show you how to remove vocals from an audio recording. I will be guiding you through the process later so just look at this movie as a simple 'taster' or for 'demonstration' purposes.

The movie is in real time. I advise that you use headphones to listen.
 
I am using an ACDC track which so happens to be 'Highway To Hell'. 

Extracting vocals/voice, or any other sound for that matter from recordings, is not an exact science. You can still hear in the background a voice singing, this is often called 'sound by-product'. It is impossible to remove everything but with EVP you don't have too. It is possible to take sound a bit further to get better results but not by much plus it can become time consuming for little in return and you often take the extraction one step too far and start to remove sound you needed to keep.

On the above movie at 10 seconds, I stop the track playing and highlight a small area. I then 'loop' (continuously replaying the selected area). On a EVP recording, which may last for a considerable amount of time, you may hear a small section that your interested in and wish to just select that one area for analyses and enhancement.

What we are now going to do is to look at two sound files, again using the same ACDC recording. The first sound file will be of the untouched audio track with a frequency analyser activated. The second sound will be the SAVED audio with the vocals removed and again placed into a frequency analyser. The intention here is to give you some knowledge for when you analyse and enhance your own EVP recordings.

For demonstration purposes I am going to use a small section of the audio which contains the vocals.
Untouched/unedited Audio.

Same recording but with vocals removed using software.

I have used the same procedure as the first movie at the top of this page.

There are clear differences between the sound files. What we have done here is to subtract the vocal band from the full sound file. What we now need to do is EXTRACT and SAVE the vocals, which is what you would do with a EVP file. As explained on a earlier chapter you can isolate any frequency using 'Riff Station' but the better results are found with a Digital Audio Workstation - DAW. I have put some thought into what the best DAW would be and come up with a FREE software Suite called 'Adobe Audition V3'. This software can be downloaded by following this LINK.

Adobe has deactivated their registration server so for registration use the following Serial Code: 1137-1004-8571-6848-7845-8029
It is common for musicians to remove lead guitar and vocals from audio to make backing or karaoke tracks. You have seen how this can be done on the above movies. We now need to learn how to isolate and save just the vocals and for that we need Adobe Audition 3. Once you have the extracted vocal (EVP) you can then enhance it further or analyse in your favorite DAW.

I am again going to use the same sound ACDC sound file and on the next movie you will see how to open the sound file and using EQ and spectrum reduction so you are left with as much vocal as possible but without the rest of the audio. As explained , this is not an exact science and with any subtraction you will loose some clarity but what your trying to do with a EVP recording which is covered with other sounds and make the voice of the sound difficult to hear, you simply want to hear the voice or sound without anything else. The movie below will be in real time which will allow you to simply follow using your own software until you become more experienced with it's use. There are a number of tutorials on youtube that will help you.

To help you further please click on this LINK to find out more about EQ Frequencies in recordings. It will help you understand what sound fits in where.

Here is the movie of importing the sound file into Adobe Audition 3.0 and applying EQ extraction and changing the frequency spectrum so we are left with just vocals or as much vocal we can. 

In this next movie I will import the raw file which was saved in Adobe into Mixcraft DAW and we can look at the frequency analyser and some basic EQ will be applied. The whole process took just a few minutes so you would be able to get a better sound file with a bit of patience but for this exercise I was simply giving you the know how.
 
 
Now go to Chapter 5.
 
googlee879f8ebce23584a