| |
Techniques
Using a Computer to Record for EVP
These are the fundamental steps for using a computer as an audio recorder for
EVP. Once you have established that you can record the phenomenal utterances in
this way, we encourage you to explore variations of this theme:
Tools
For Using a Computer
You will need a personal computer; however, it need not be a very powerful
one.
-
The computer must be equipped with a microphone
connection.
-
The computer must be
equipped with a computer application that simulates a tape recorder. There
are a number of good ones. (Audacity
is a free (Open Source) program that is very effective for EVP.)
-
You need a microphone that is compatible with a computer audio card. If
you experience a buzz when trying to record with a microphone designed to be
used with a standard audio recorder, you may need to purchase a microphone
specifically designed for a computer. Some experimenters use a splitting
cable that allows two mono microphones to be plugged into the microphone
input to the computer. They record in mono and place one microphone near a
computer cooling fan for background noise and the second on the desk top to
record questions.
-
About 100 Meg of extra hard disk memory for audio files
is suggested and a way to transfer saved audio files to CD is a
plus.
Sound
Source
You will want to have a sound source that can be placed near
the microphone. Research has shown that the communicating entities use available
audio energy to form their words. EVP Experimenters have found that the
entities will actually remodulate ambient sounds as they are picked up by the
microphone, and within the electronic circuitry. Computers usually have a
relatively quiet recording circuit, so it is recommended that background sound
is provided during the recording process.
The screen print of the audio editor,
Audacity, shows a sound file containing words spoken by the experimenter
and the EVP, and the relatively volume of each compared to the
background noise. Use this as an example of a useful loudness for
background noise.
Here is the audio file:
Betty's
in there
Almost all attempts to improve the
Quantity and Quality (QQ) of EVP are different forms of energy
modification, as opposed to different forms of psi detection. People try
every imaginable form of sound in an effort to record more EVP that are
more easily understood. Here is a guide for the selection of audio
noise:
-
White noise was used in the early days of EVP
because the only electronic device available to people for noise generation
was the radio turned off-station. That produced various versions of what is
known as "white noise;" however, the human voice is mostly in the range of
200 to 4,000 hz. Since most of the audio energy in white noise is not used
for voice, any sound source more tailored to human voice is better for QQ. A
common household fan, for example.
-
Amplitude modulation (as in AM radio) has been
found to be a little better as a sound source than frequency modulation (as
in FM radio). We think this is true because natural voice is made of
variations in both frequency and amplitude, and FM sources of sound do not
generally have amplitude variations. Digitally generated noise is not as
effective as "live" noise found in the environment. For instance, we hear of
more QQ from running water than from digitally generated white noise.
-
Use of more than one sound source will cause
layering of EVP. For instance, we once used a small water fountain which
produced a tinkle of noise, along with a large household fan which produced
more human voice frequencies. The resulting recording had EVP that sounded
like normal voice in tone with elfin-sounding voices present at the same
time. This is very distracting and points to the need to manage background
sound to optimize human voice frequencies.
-
Some inexpensive digital voice recorders have
relatively noisy audio circuits. We think this is because of the low sample
rate and resulting first significant bit error due to small word size of
each sample. See
A
Testable Hypothesis Explaining EVP Formation in Digital Recorders
for a possible explanation. The noise produced in these lower quality
recorders is excellent for the formation of EVP, making digital voice
recorders very useful for field recording. The noise from cassette recorder
motor drives turns out not to be useful for EVP, making an external
microphone necessary for these recorders. Also, the tape hiss is noise
produced at the wrong part of the electronic circuit for EVP voice
formation.
-
A recording of crowd babble that has no distinguishable words will work.
ATransC offers a Portuguese language, crowd babble sound track that has been
made available to us by Sonia Rinaldi.
We do not recommend the use of any broadcast noise in which discernable
voices of any language are present.
Using an External Audio
Recorder
You can record into a stand-alone audio
recorder and then play the results into the computer for listening and
editing. This is a good way if you
intend to archive all of your recordings, because a typical two-minute recording
can produce a computer file of two to three megabits. You can store the audio file on
the tape rather than in the computer; however, we prefer storing files on
a compact disk.
Recording
Session
A typical recording session is about two minutes. You will
want to listen very carefully to every click and syllable you record—sometimes many
times, so two minutes is just about right.
-
In
Adobe Audition, under FILE, selecting NEW and
then Sample Rate = 11025, MONO and 16-BIT. This will produce a
2.6 Meg file at two minutes recording time.
-
In the conversion of the analog sound of your
voice to the digital sound in a computer, faithful reproduction of the
analog sound requires at least two samples per cycle. We can hear up to
around 22,000 Hz, but human voice is pretty much below 4,000 Hz. Therefore,
a sample rate of 8,000 Hz will faithful reproduce the majority of human
voice frequencies. This has traditionally been the practice of telephone
companies. We recommend that you use a sample rate of at least 11025 so that
you can preserve some of the higher frequencies, which tend to help voice
recognition. There is sometimes a problem with converting *.wav format to
mp3 when a lower sample rate is used.
We use 22050 stereo because we like the fidelity and because we like the
sound in both ears. A 32 bit digital word, as opposed to 8 or 16, seems a
little overkill.
-
Essentially the same instructions apply to
Audacity. See Setup for
Audacity for instructions.
-
Follow the instructions at
Basic EVP Recording
Techniques for your initial recording effort.
Recording directly into a computer for
EVP can be very effective; however, it is usually necessary to add
background noise because the computer sound circuit is usually music
quality.
An
Experiment
Since a computer is capable of
supporting stereo sound, the use of a single microphone in stereo mode
offers some interesting possibilities. For instance, theory has it that
the words in EVP are formed in the electronic circuit, and that no two
recorders will (as a rule) record the same EVP at the same exact time.
When being used in the stereo mode, a computer is technically two
recorders--left and right channel. In principle, it should be possible
to record using one microphone across the left and right channels, and
then to invert one channel and average the two channels in a audio
management program. The result should be zero signal as the two
identical signals from one microphone cancel one another out. If an EVP
is present, it should be on only one of the two channels, and therefore
survive the cancellation. If this is true, then this may be an effective
way to recover the EVP from the noise.
This could be accomplished in real-time to reveal the voice as it is
spoken. The problem here would be in recording Class A or at least
understandable words.
One technical issue
in this is that the left and right channels of a stereo input to
computers are not always well balanced. In other words, the left channel
may have a little more background noise than the right, or perhaps a
little more zero off-set. It is not clear how to manage this, unless a
pair of variable resistors (volume controls) can be places between the
microphone and the splitter. We have not
had the time to test this approach, so we must depend on the ingenuity
of people such as yourself to see if this is a valid approach, and if
not, why. We always need to know if an idea fails and try to learn why,
so please let us know via
email. |
|
|
| |
|
Community Involvement
Help improve these articles
ATransC is a publicly supported
organization. Our mission is to bring this information
to the public as clearly and correctly as possible.
ATransC Members do
all of the heavy lifting when it comes to financing the
operation and making this website available for you.
You can help by letting us know if you
find a typo or something that grammatically does not
make sense. Use the comment tool where provided and let
us know how we can improve articles.
We are always happy to receive
constructive input.
|
|
|
|
atransc2009-banner.gif){kind=spacer}
