|
|
|
|
THE REVERSE VOICES PHENOMENON
It is very easy to understand the meaning of "reverse" voices. Let’s say that when we record, the tape recorder rotates in a certain direction, which could be called "forward". The spiritual communicants possess the ability to record in the opposite direction, and those cases constitute the most contusing evidence of the phenomenon’s authenticity.
The first transcommunicator to observe that spirits could record "forwards" as well as "backwards" was Jürgenson.
To do the reversion with a cassette tape recorder is a difficult task. If the experimenter works with reel tape recorder (like the American pioneer that still uses this kind of equipment) it is much easier, because he only needs to invert the reels. But cassette tapes are complex, the cassette must be opened and inverted manually. However, those who have a computer can make this operation in fraction of seconds.
To give a real example, let us say that in a recording was registered:
- "Friends, are you hearing me?"
If we give the "reverse" command, commonly found in recording software, immediately this sentence will change to:
- "?em gniraeh uoy era ,sdneirF"
Obviously, when we invert a recorded sentence and listen to it, we can not understand anything.
WARNING – BE CAREFUL!
At least three cases have already been sent for us to analyse. The experimenters had found recorded sentences that seemed to be spoken in "unknown" languages, so they deduced those were paranormal recordings.
It happens that, in 95% of the cases, those occurrences result from the use of neglected twisted tapes. Therefore, there is no phenomenon.
As an example, we can quote the case of an associate who sent us a tape with about five minutes of this: nhumbtijlbh, nhfyb, hnmjgppk; etc.
Unknown language? Reverse voice? What phenomenon could that be?
We sent the tape to engineer Valdir Cunha who carefully inverted it manually, which was a hard job (at that time he did not use the computer for audio work yet). At first, he copied the cassette tape recording to a reel tape recorder and finally he wrote down the sentences one by one.
After several days working on the case, he concluded the recording consisted of a spiritualistic poetry.
I was very glad and sent the result to the associate, so she could study the case. Immediately after she had listened to the tape, she said:
- " But that is a recording I made of a spiritual medium’s speech in Uberaba, at the end of a meeting in which I participated... and it is also his voice.
Well, once the medium was not dead and the message was exactly like the original recording, it could only have happened something very simple, that finally was confirmed. When making a copy of the recording, the tape accidentally coiled inside the recorder. She had not realized the tape had been twisted and ruffled. The audio was registered in the tape’s reverse side. That tape was put away and nobody recalled of it anymore. After some time, the associate wanted to reuse the tape and make a new recording. Then she was surprised when she heard the "unknown" foreign language.
It was easy to check this occurrence: we took the original tape and rolled it manually until we found the twisted part of it that caused the "false phenomenon".
We share this case with our readers because it happens more often than anyone can imagine.
A TRUE CASE
We are trying to warn the reader that it is important to be very careful. The phenomenon has to be very well studied, in order to maintain its credibility.
On one hand, we can come across cases like the one described above (of neglected tapes). On the other hand, there is also the possibility of a true phenomenon.
In January 1998, a scientist came to our "station" in order to adjust our image reception’s equipment. While Dr. Augusto gave us several explanations about Physics, unpretensiously we decided to record the conversation. Among several subjects, we spoke about the impressive reverse voices, and how, technically, could it be produced opposite to the Time.
After his departure, I listened to the recording and there were, as always, dozens of comments made by the spiritual friends who had heard our chat. Then, I thought it would be interesting to try to listen to the recording’s "reverse" mode.
Surprisingly other dozens of audios were there. But, three cases let us astonished, because the spirits had modulated Dr. Augusto's voice, which means that we had an absolutely irrefutable case of the true phenomenon of reverse voices.
That is, when playing the tape in the "normal direction", we can listen to Dr. Augusto's voice saying something and if we invert his voice, instead of listening to the incomprehensible and typical "nhvtomp ngrendjl mhntyn", there is another complete and understandable sentence and the voice is different from Dr. Augusto’s!
(The reader can check for himself. The recording can be copied to the computer and then reverted. There will be another sentence instead of incoherent sounds. This is the unique case we have knowlegde, of reversed voices over a HUMAN voice).
HOW IT HAPPENED:
When Dr. Augusto says in his explanation:
- "Everything
is going... to be distorted. It will be in such a way..."
Portuguese:
- "Tudo isso vai...vai distorcer. Vai ficar numa maneira tal..."
[In the following
technical report, this recording is given the code TR1]
 |
In the reverse mode this changed to:
- 1st voice: "Put
it there. Let’s give a look".
Portuguese:
"Bota aí. Vamos Olhar"
[In the report,
the code for this sentence is Tst1]
- 2nd voice: -
"I wanted to see if you are my uncle"
Portuguese: "Queria ver
se você é meu tio"
[In the report,
the code is Tst2]
|
The second case:
Dr. Augusto says:
-"(...) up to this point I can really understand what is being said... these are the transformations (..)". [in the report this is Tr2]
When reverting the whole sentence, part of it is not understandable but there are also two clear voices that modulate Dr. Augusto's voice and say:
- "I am José" and "What a pretty young woman". [In the report, this is Tst3]
And the 3rd case, taken from that night’s same recording:
Dr. Augusto said while he made drawings and graphs in a piece of paper:
-" (...) Sônia, here is this one, right?
In the reverse mode of his voice, it says:
-"(...) Engineering from here". [In the report, this is Tst4]
Note: Apparently, the explanations given by my friend checked with the engineering from Beyond.
The fact that the recording is loud (since it was the own human voice modulated in the reverse sense) will be good as reference to any unbeliever.
It is impossible to "manufacture" a voice that when reverted changes into another different sentence.
Note: To imagine that inverting the position of the letters of a certain sentence could result in another correct and complete sentence, it would be more difficult than to win the lottery’s prize three times in a row.
SUPPORT OF THE SCIENCE
We sent samples of the audio above and requested to be investigated if Dr. Augusto's voice and the reverted one were, by any chance, the same. This would represent a step forward in the confirmation of paranormal interference.
Here is the report, strictly elaborated within internationally accepted patterns and entirely transcribed below:
Pieces of Examination:
For this study there are three motivation-pieces which are three recordings numbered 1,2 and 3, recorded and put into digital form by the use of an IBM-PC computer, by means of a standard sound board, namely, Sound Blaster. The used resolution was one with 16 bits, with a sample rate of 11025 samples per second.
Objective of the Examination:
To verify that all speeches have been pronounced by the same speaker, from now on named speaker A.
Standards of comparison:
6 parts of the recordings are objective of standard comparison and are numbered as follows:
· 1 part which lasts for about 4.8 seconds of Speaker A's voice, here named trl, taken from recording 1.
. 1 part lasting about 2.3 seconds, of Speaker A's voice, here named tr2, from recording number 3.
· 2 parts of a recording of an unknown speaker, here named tstl and tst2, taken from recording 1.
· 1 part of the recording of an unknown speaker, here named tst3, from recording number 2.
· 1 part of the recording of an unknown speaker, here named tst4, from recording number 3.
Pre-processing of the Samples
The samples have been standardized so that they have the same signal level and the same signal-noise relation. The parts that have no voice have been removed, as they do not characterize the speaker. Out of the standardized samples there have been taken Mel-Frequency Cepstral Coefficients - MFCC, using windows of 23.2 minutes, with spacings between them, of 10 minutes, 5 filters per octave, resulting in 16 coefficients per window.
Methods of Comparison:
The coefficients of the samples tr1 and tr2 have been used for experimenting an automatic recognizing system of the speaker, independent of the text, based in Gaussian mixtures (GMM- Gaussian Mixture Models), using 24 mixtures to represent the speaker.
As an algorithmic of comparison it has been used the logarithm of the verisimilitude (loglikelihood). This method has been chosen because of its strength, but it is not the only one that can be used.
The attained
results are listed in the table below:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The number of vectors indicate the number of windows resulting from the pre-processing, the output value is the sum of the probability logarithm of the vector to belong to the concerned speaker, and the standardization is done by the division of the output value by the number of vectors.
Conclusion
By means of the obtained results one concludes that there is no evidence that all phrases have been spoken by the same speaker, because the standardized values are not close enough.
Our Comments:
This case is one of the most impressive that we have experienced, because Dr. Augusto's voice, being inverted, generates another voice, although it was his own voice.
Regarding the identification of the four paranormal voices [tst1, 2, 3 and 4] the values suggest them to belong, at least, to two different entities.