• Home
• ATransC History
• Meet The Directors
• Support the ATransC
• Contact ATransC
• Search

• About Membership
• Online Membership Form
• Idea Exchange
• NewsJournal Archive

• Introduction
• Psi Phenomena
• About the Big Circle
• Physical Mediums

• Introduction
• White Paper About EVP
• ATransC Survival Hypothesis
• Terms

• Introduction
• Audio ITC (EVP)
• Visual ITC

• Introduction
• Transform EVP
• Synthesized Voice EVP
• Photographic Phenomena
• Video Phenomena
• Other Techniques

• About
• Questions About Evidence
• General ITC Questions
• EVP/ITC Techniques
• Concepts
• Equipment and Software
• Researching Transcommunication
• Sarah Estep Research Fund
• History of TransCommunication
• ATransC Member Benefits & Services
• Etheric Studies

• About
• Idea Exchange
• NewsJournal Archive
• Books and Videos
  • ATransC Books
  • ITC Books
  • Books on Survival
  • Etheric Studies Literature
  • Video of Interest
  • Creating and Reading an eBooks

• Member's Websites
• Circles
• Sister Organizations
• Related Subjects

Editing Sound Files

The Objective

We describe an EVP message as being:

Class C: Poorly heard, even with headphones; probably not understood by most listeners

Class B: Usually requires headphones but sometimes heard over speakers, and understood by some people,

Class A: Clearly heard and understood by the average person when heard over speakers.

It is possible to enhance a Class B or C EVP with sound editing software so that it can be understood. The objective of such enhancement is to selectively amplify the sound track, reduce noise and possibly adjust the speed of the voice. Using Audition as a model, we recommend that the first step is to perform a SAVE AS function using a different title for the sound file to assure that the original is not lost. Make sure the ENABLE UNDO option is selected in the EDIT menu. Then select all or part of the sound track, and under the TRANSFORM menu item, begin trying various feature, depending on the condition of the track. You cannot hurt anything so just try different things to see what happens. Always use a headset.

Assuming the suspected EVP is hidden in the background noise of the sound track, begin with the noise reduction function that is described below. Try this at 50% reduction and then at 75% and 100%. If this does not help, UNDO and then try the FFT filter under the TRANSFORM, FILTERS, FFT FILTER. There is a PREVIEW option in that window so you can experiment and listen to the results. Remember that you will usually want to remove most of the sound above 3000Hz so you might want to begin with one of the NOTCH filters.

Assuming the filter helped, try the noise reduction step again. Did that help this time? If not, then UNDO.

You may want to AMPLIFY the signal at this time using the AMPLITUDE features. If you can hear words but they seem to be spoken fast or slow, try using the STRETCH function under the TIME/PITCH feature.

The point is, there are a number of tools available to you in sound editing software. Which ones you use and how much they are applied depends on your particular situation. Enhancing software to make an EVP clear is something of a craft. The only way you will learn that craft is to experiment.

Research reported by Paolo Presi and his coworkers has shown that EVP are formed as a "thickening" of the available noise. A normal voice is formed from a suite of frequencies derived as the voice box frequencies pass through the mouth. (See also Computer–Based Analysis of Supposed Paranormal Voice) After conversations with Presi and Alexander MacRae, we believe that we are culturally trained to use a number of cues based on frequency relationships to help us understand words. Since the EVP does not normally have the usual frequency relationships, it is common for different people to interpret the same EVP in different ways, even though the samples seem to be "obviously a Class A." The message to you is that filtering and noise reduction can change the EVP so that the meaning is changed. This poses a good argument for conservative manipulation of sound files and caution in concluding the meaning of EVP.

A word about saving and sharing EVP samples is in order here. ATransC members usually use the ".wav" format for saving and editing files, but mp3 format for sharing and posting on the Internet. If you intend to share the EVP sample with other members, we suggest that you keep the file small so that email servers will not reject the attachment, or that you arrange to post it on a personal website and simply send the address to other people.

Noise Reduction for a Sound File

We have found that, if you have a relatively consistent background noise in your EVP recordings, say from a noise generator or from a fan, you can use Audition to effectively remove that noise. This is based on an early version of Adobe Audition, so you should think of it as more of a set of generic instructions that can apply to most sound editors.

To begin, open Audition and start a fresh recording session: Under menu item, FILE, click on NEW and set Sample Rate = 11025 (or better), Channels = Mono, Resolution = 16 bit, then click on OK. (You need a sample rate that is at least twice the frequency that you want to listen to. For instance, telephone lines are rated for 4,000Hz and have a digital sample rate of 8,000Hz. The recommended sample rate of 11,025 should faithfully reproduce 5,512hz. Virtually all of our EVP are under 3,000Hz. The lower all of these numbers, the smaller the file size. If you have trouble loading a sound file with one sample rate, try it with another.)

Play your sound track into the computer while Audition is in RECORD mode. (With modern recorders, you will normally transfer the file from the recorder to the computer via a USB cable and then open it in the sound editor.) See Recording Sound into a Computer.

For noise reduction, once the sound track is loaded into your computer, select a few seconds of sound track that has typical noise but no voice (or no suspected voice). Go to menu item, TRANSFORM and select NOISE REDUCTION and NOISE REDUCTION again. Click on GET PROFILE FROM SELECTION. (If you do not have a large enough data sample selected, this option will not be available.) You will notice that a graph will be displayed and SAVE PROFILE will become available. You can save the profile and use it as a "standard" filter for similar background noise.

(If you cannot find a large enough portion of your sound file to sample without getting suspected EVP, sample what you can, create a new file and past that sample several times in a new file to create a longer file to sample.)

When the application has finished, click on CLOSE (do not click on OK) and then select the entire sound file or that portion you wish to analyze. Then open the noise reduction window again and click on OK. This will filter the selected sound track based on the profile of the previously selected noise. (You can undo this if you do not do a SAVE function first. Be sure that you have ENABLE UNDUE checked under the menu item, EDIT.

Since the EVP is usually formed out of the background noise, reducing the noise may reduce the voice as well. Try different NOISE REDUCTION LEVEL settings. This is a slide selector just under the noise profile graph. This may be an iterative process to find the best combination.

Depending on the uniformity of the noise on the track, the voices should stand out more from the noise. Please note that this process sometimes induces a "ringing" sound into the sound track. This noise is considered an artifact.

Filter

You can also work with the FILTERS option under the menu item, TRANSFORM. The FFT Filter will permit you to listen to the sound track while you adjust the graph to find the best setting. Do not be afraid to us AMPLIFY and TIME/PITCH while you are trying to figure out the EVP.

Remember, your first interest is to make the EVP understandable. It is best to make sure that any EVP that you wish to demonstrate to a friend is understandable and Cool Edit will help you do this. However, some people will complain that you have changed the sound track, and have therefore, disqualified the EVP. This may be true for scientific inquiry but it should not be an issue for simple demonstration and personal use.

Over Processing

As we learn more about EVP, we also learn about what people hear and how they make mistakes with EVP. The articles, Objectivity test for EVP and Characteristic Test for EVP will give you a sense of how people hear EVP and common characteristics to look for to recognize EVP. Here are a few points to keep in mind as you begin working with sound files:

• There are thousands of EVP, but relatively few Class A examples. The average person with little experience listening to EVP will typically correctly understand only about 25% of the words of a Class A example. This means that you should consider sharing only the Class A and be prepared to guide your listener.

• Think very carefully what you are going to do with your EVP examples. There is little sense in saving everything.

• This is communication and the entities speak to us where we listen. Virtually all of the techniques are either novel ways to modify sound for voice formation or ways of enabling intentional influence of random processes to determine sound selection. Indications are that the Transform EVP requires an analog stage for formation, so if you think EVP are only formed in the very high or very low frequencies, remember that you must translate that into audible frequencies and virtually always pass it thorough an analog stage where the voice is probably being formed.

• We do not recommend looking for EVP on the reverse path.

• In current studies, people hear voices where there is only noise about 50% of the time. While they usually report not quite making out what is said, some people are able to find very elaborate utterances. All of us are subject to this problem. If an EVP cannot be correctly understood by a few friends without prompting, discard the file. See EVP Online Listening Trials and Phantom Voices.

• Story telling is when the practitioner records until a likely sound is recorded and then creates a story to make the sound seem like the correct answer or a meaningful comment. It is very common and another reason that a listening panel is so necessary.

• There is analog radio just about everywhere. Hand-held devices tend to act like antennas collecting that radio signal. It is virtually always present in your recordings and amplifying it very much can make that RF contamination audible. We recommend no more than about 20 db of implication.

There is much more to consider, and we hope the rest of the website will fill in the gaps. Please feel free to use the contact tool to add ideas, make comments or help us with types.

Donate to ATransC   |   Pledge   |   Top   |   Contact the Association TransCommunication

Unless otherwise noted, this work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License