Introduction:

The EVP Project page will serve as an experimental platform for the independent analysis of a selected group of EVP recordings.

For this project, EVP will be loosely defined as "Any audio recording that one believes to have been caused by a paranormal force exhibiting its presence via electronic media." The actual recording method used may vary from one researcher to another and is not limited to any particular "standard model". You're invited to send a few of what you consider your best examples of EVP recordings to be included in this project.

Your audio files should be sent as email attachments with the following stipulations:

1) Standard WAV files are preferred. Please save your files in WAV not MP3 or other format. (Why WAV? See here.)
     The audio should not be computer enhanced in any way, no effects editing allowed!
     Also, very important:  Please include at least a few seconds of background audio that came ahead of and
     after the occurrence of the EVP. Don't send a clip so tightly trimmed that only the suspected EVP passage is heard;
     some background noise sampling must be included to establish a baseline or "null" comparison.

2) IMPORTANT: Do not give your files a name that reveals what you believe the EVPs are saying.
    Instead, please give your files a numeric name or simple identifier such as Smith_01.wav,
    and again, do not disclose in the same email what you think the EVPs in your recordings are saying.

3) Use a separate email to tell me what it is you think the EVPs are saying, for example:
    Smith_01.wav = "That's a good idea."

4) In the second e-mail you should also provide details re the circumstances of each recording: The location and
    setting, date and time, and a description of the methods used to obtain the EVPs. Include the type of recording device,
    type of microphone (or other method?), mention how many people were present, plus any other relevant factors.

Please limit your submission to no more than 3 audio files. When I feel that I've accumulated an adequate and representative sampling, I will post a selected few (probably 8-12) of the files to this site, and then invite the public to render their opinion as to what the voices might be saying, and to offer their judgment re the quality and validity of the recordings, etc.

Project results will be posted and discussed at an appropriate future date, depending on the response.

Please send your EVP samples with this subject heading: EVP PROJECT: Samples
To:  jim@gobeyondnow.com               Or simply click here: EVP PROJECT: Samples

Please send the associated information with this subject heading: EVP PROJECT: Details
To the same address as above ....................................or click here: EVP PROJECT: Details

When I receive your samples and details I will reply to acknowledge them, usually within 24 hours.
 

I look forward to your emails and can't wait to hear what develops!

*          *          *

Why WAV?

WAV, short for Waveform Audio Format, was originally developed for use with Windows operating systems, but these days it is compatible with virtually every computer-based audio recording, playback and editing platform; it is one of the most commonly used formats for converting audio information into digital files. WAV files are typically preserved in an uncompressed and "lossless" manner which, without getting too technical, means that a WAV file is as close to matching the original source's quality as possible. (Within the parameters of the capture methods and settings being used that is, I'll discuss those issues later.)

Quoting from the Wikipedia:

[WAV] is a relatively “pure”, i.e. lossless, file type, suitable for retaining “first generation” archived files of high quality, or use on a system where high fidelity sound is required and disk space is not restricted.

The only downside of WAV files is that they can become quite large and thus take up lots of space on a disc or computer's hard drive. In order to reduce the amount of storage space required, several alternative formats have been developed that utilize audio "compression" schemes. MP3 (Motion Picture Experts Group Audio Layer 3) has become one of the most popular alternative formats for audio file storage and sharing; the MP3 encoding process will typically compress the original audio by a factor of 10 to 14, meaning that the resulting file size would be just a small fraction of the equivalent WAV file. Obviously, this gives MP3 a major practical advantage over WAV, but the savings in terms of storage space comes at a cost in terms of quality.

MP3 is by its nature a "lossy" encoding format, meaning that a certain amount of the original source information will be lost in the process of creating an MP3 file. Basically, what happens with MP3 is that certain frequencies will be determined unnecessary for the overall audio experience of a given passage, and those frequencies will be eliminated when the source material is processed and the end file created. (The "lost" frequencies cannot be recovered by re-converting the MP3 file to a different format.)

Quoting from the Wiki again:

The use in MP3 of a lossy compression algorithm is designed to greatly reduce the amount of data required to represent the audio recording and still sound like a faithful reproduction of the original uncompressed audio for most listeners, but is not considered high fidelity audio by audiophiles.

What all this boils down to is that typically, as much as 90% (or more) of the information that was available with the original source audio will be lost after encoding to MP3 format! The rationale is that this 90% was not essential to the listener's audio experience, and that the missing information will go unnoticed by the average listener under normal circumstances.

OK, discarding 90% of the information that was originally available in a musical selection might be acceptable for casual listening purposes. Personally, I think 100% would be even better for most of what passes for music these days, but do we really want to throw out 90% of the information that was produced during the occurrence of an EVP event? I don't understand why we would want to discard any of that information, and if you consider EVP to be a subject worthy of serious research, I don't think you should either. After all, the frequency elimination algorithms used in the MP3 encoding process were developed for music listening, not for EVP research purposes. 

Using MP3 might not make a big difference in what we actually hear when we listen to an EVP recording, but there is no question that it will make a very big difference in terms of what was actually produced by the original source! Why in the world would we not want to have the "purest" and most accurate recordings possible with which to perform our EVP analyses and render our judgments?

I'd like my EVPs straight up, raw, whole, and uncompressed. Not shaken or stirred, thank you very much.

Look folks . . .

Given the fact that EVP recordings are typically just a few seconds in duration, the question of file size is hardly a serious issue with today's computer storage capacities and high speed transfer rates. Since WAV is a lossless and currently the most universal format, I can't think of any good reason to capture, store, or share your digital EVP work in a format other than WAV -- at least not until something else comes along that really is better in terms of audio quality and compatibility with EVP research.
 

Compromising the quality of your audio recordings for the sake of saving a little space
on your hard drive is an indication that you didn't think the files were worth saving in the first place!!!
 

*          *          *

References:

    The Wkipedia article on WAV quoted above
    The Wikipedia article on MP3 quoted above

Here's a good introduction to How MP3 Files Work with a graphical explanation of its audio compression scheme.

Here's an excellent Primer on Audio Formats that discusses the advantages and disadvantages of various
encoding systems, including one called FLAC that may be an up-and-coming contender for replacing WAV.

Here's a more technical discussion of Digital Audio Best Practices from the CDP Audio Working Group.
(It's a 60 page pdf document and is well worth a few moments of your time if you're really serious about recording EVPs.)

Suggestions on Choosing a Sampling Rate:

Any audio recording is, at best, an approximation of the original sound source. How close the approximation comes to matching the original depends on many factors, including the quality of the input device used (generally a microphone for EVP purposes), the design characteristics of the recording device itself, the execution of proper recording techniques, and in the end of course, the quality of the playback equipment used.

If you are using hand-held digital voice recorders with their built-in microphones for your EVP experiments, you are not going to get very high quality recordings. One variable you should be aware of to ensure that your recordings are at least as good as they can be however is the "sampling rate". Most digital recording devices and computer recording software let you choose the sampling rate, this determines how many times per second a sample of the audio source is taken. The higher the sampling rate frequency, the more samples taken per second, and the higher quality your recording will be. A higher quality recording results in higher demands on storage space for the completed audio file though, so a compromise is often made between quality and quantity.

The general rule of thumb is that your sampling rate should be at least twice the frequency of the highest audio pitch you expect to record. Thus a sampling rate of 8,000 Hz (8 kHz) is very adequate for recording voices and would be appropriate to use at lectures, and for making verbal notes to yourself, etc. Since you may think of EVP as being an exercise in voice recording, you might be tempted to use 8 kHz as your sampling rate. (11,025 Hz is the next standard step up, and this may be the lowest setting available on some recorders.)

In fact though, using these lower settings will result in lower overall recording quality and when used in EVP work this could cause a sort of "audio blurring", especially with the higher pitched sounds that might occur in the environment. For example, the high pitched squeaks of a door being opened might not be recorded very accurately and the door could end up sounding remarkably like a voice when you hear it played back. The question is, shouldn't you strive to make your EVP audio recordings the best that they can be?

If you're using a digital recorder's built-in mic, or an external mic of el-cheapo variety, the highest frequency it will be capable of faithfully recording is probably in the neighborhood of 10,000 Hz. Therefore, a sampling rate of 22,050 Hz would be the lowest one to choose for best file size-to-quality combination. If you are using higher end recording equipment (and this can include computers as well as stand alone recorders fed by good external microphones) the sampling rate should be set to 44,100 Hz .

If your recording system allows you to set the bit rate (or bitrate), 16 bits is generally recommended as the minimum for good quality. CD quality is 44,100 Hz at 16 bits. Some professional recording gear allows even faster sampling rates such as 96,000 Hz, and higher bit rates, with 24 bits being fairly common. These settings may be overkill for our purposes since the recording quality would probably exceed the limits of our hearing, although such super-quality recordings might have definite research potential.

Also remember, using a "lossless" uncompressed format such as WAV (or AIF, FLAC, etc.) should be the standard of choice for capturing and storing your files for serious EVP research. The sampling rate should be set to at least twice the highest frequencies that can be captured within the specs of the equipment being used, and the bit rate should be set to at least 16.

Compromising the quality of your audio recordings for the sake of saving a little space
on your hard drive is an indication that you didn't think the files were worth saving in the first place.

On Evaluating EVPs:

The goals of Go Beyond Now's EVP Project are 1) To attempt to establish an objective judgment as to whether any such phenomenon actually exists, 2) If it does, then to try and understand the phenomenon better in terms of its causal properties, and from there, 3) To devise improved methods for obtaining and recording the EVP messages.

Earlier in the EVP Project page, I presented a working definition for EVP as being: "Any audio recording that one believes to have been caused by a paranormal force exhibiting its presence via electronic media." The word believes in that definition applies in terms of choosing samples that may be subjected to our tests of course. Belief in EVP as a phenomenon does not make it a fact, although many people in this field of interest seem to believe that it does.

What standards can we apply that will help to objectively determine the existence of EVP as a genuine phenomenon and that it is not just the product of our very human, and very well-documented tendency to think we are hearing voices when the only thing we are really hearing is unintelligible noise?

The first aspect of The EVP Project will be to try and establish whether or not there is a statistically significant agreement among listeners as to the message reportedly being presented by an EVP voice, or for that matter, whether it is generally agreed that any verbal message at all has been presented in the case of each test sample.

Although agreement as to the perceived messages would not in itself constitute "proof" for EVP, it would be a substantial first step toward justifying the continuation of this line of research. Unless people can agree on what an EVP voice is saying, or at least that there is a voice, why should we even bother to waste our time, money, and energy in the pursuit of EVPs?

On the other hand, if no agreement as to the perceived messages can be established, this would not "disprove" EVP, but it would certainly call its validity into serious question, and at the very least it should introduce an element of objective doubt in the minds of truly open-minded observers. What I mean here by "objective doubt" is that with this project, the opinions are expected to be rendered primarily by other paranormal researchers and EVP believers, not weighted toward the side of those who are primarily skeptics, which often becomes the case in such studies. In other words, our test is admittedly expected to be heavily biased in favor of the EVP believer, and therein lies its significance: any negative results should be viewed with special interest to other believers since they won't be able to call foul on the part of the usual crowd of chronic debunkers.

For this particular project, you should take particular notice of the fact that we've asked for recording samples to include a few seconds of the background noise that comes before and after the occurrence of the actual EVP. This is seldom done with the presentation of EVP samples one encounters on the internet, but it is absolutely essential in order for the listener to establish a reference point by which they can judge the audio characteristics specific to the EVP itself.

We've also asked that the samples being submitted for inclusion not be processed by any audio editing effects, the samples should be as close to the original "pure" audio source as possible. The importance of this for making objective analyses should be obvious, but more and more, people seem to be presenting their samples on the web in post-processed, and therefore favorably biased form.

Another key factor is that samples will be presented without a prior suggestion as to what the original observer believes the words to be saying. Again, the reasons for this should be completely obvious for those truly interested in objective research, yet it is exactly the opposite of how EVPs are presented on every single public EVP forum I have ever come across.

By introducing these few significant differences in the way our EVP samples will be presented, and by virtue of the fact that we expect the judging to be done primarily by those who are "favorably biased" toward the existence of EVP, we hope to produce results that will be especially meaningful and universally acceptable to the paranormal research community.

To be continued . . .

The material on this website (except where otherwise indicated) has been prepared by J. Hale.
All original content is copyright protected and all publication rights are reserved. Effective September 1, 2006, and beyond.