Types of EVP
This is a rapidly evolving field of study, and our reported understanding of what EVP are and how they are formed should be considered “what we think today,” rather than: how it is.” The voices in EVP are formed in three very different ways, but both are dependent on supplied audio-frequency energy.
Traditionally EVP formation has involved the transformation of available audio-frequency energy into voice, which is thought to occur in the electronic equipment. The resulting signal is seen as a simulation of human voice which may very closely mimic the physical voice of the person thought to be speaking. This includes nuances of voice, such as accent, age, sex and attitude. Analysis of the resulting voice usually shows novel arrangement of formants (frequency grouping by octave of the voice box frequency developed during passage through the mouth35) and fragmented voice box frequencies (Formant 0).13 Transform EVP was traditionally accomplished by using radio static (a readily available source of sound in the early days of EVP study) as background sound. Current Best Practices involve the use of unmodulated noise, such as supplied by a fan, but most EVP are recorded today using a digital voice recorder, and the device tends to provide ample noise for voice formation during normal operation.
This depends on a random process which is thought to be influenced by the communicating entity. In the application known as EVPMaker.36 A pre-recorded sound file containing voice is stored in a buffer, and then a random process selects segments of the stored file from the buffer to produce a new audio file. Recorded human speech has been traditionally used, but the AA-EVP has discouraged use of any form of “live voice” for EVP experimentation. However, EVPmaker developer, Stefan Bion, has recently provided a sound file containing speech fragments known as allophones, which have been generated by a speech synthesis program. In this application, if a word is present in the output, it must be “fortuitously formed” by a chance arrangements of allophones or it must be the product of intended manipulation of the random process. The deciding factor is whether or not the utterance is meaningful for the circumstance. An example of this is at real-time, two-way conversations.
Environmental control of speech synthesis:
A new approach to EVP has been the use of environmental energy sensors to control the operation of a speech synthesis process. In the Paranormal Puck,37 this is accomplished by sensing environmental electromagnetic, temperature, magnetism or electrical changes around the device. The device connects to a computer via a USB cable and the computer has a supplied program that uses the sensor information to control a micro-chip in the peripheral device to produce voice. See real-time, two-way conversations for an example of this form of EVP.
What is probably not EVP:
We are working on an article titled EVP Formation, intended to help explain what we know about EVP today. An important part of that article is a discussion about common factors. Based on those common factors, and the result of other studies, we have begun to feel confident in saying that certain technologies (probably) do not produce EVP. The technology that comes up most often is radio sweep, and in a case study, we report that it probably does not produce EVP as it is reported. There are clearly instances in which the noise produced by rapidly sweeping radio stations is used to produce transform EVP when the sweep output is recorded. The radio sweep process may be beneficial as an aid to the operator's intuitive understanding of the question.