Links Andere Webseiten zu Tonbandstimmen und verwandten Themen
Using EVPmaker with SpeakJet Allophones
This article explains how EVPmaker can be operated with SpeakJet Allophones
by S. B.
This article shows how EVPmaker can be set up to use the 72 allophones of the
chip as an audio source to generate a robotic-like random voice which seems to produce interesting results when used for EVP sessions.
Excerpt from the SpeakJet User's manual:
"The SpeakJet is a completely self contained, single chip voice and complex sound synthesizer. It uses
Mathematical Sound Architecture™ (MSA) technology which controls an internal five channel sound
synthesizer to generate on-the-fly, unlimited vocabulary speech synthesis and complex sounds. The SpeakJet
is preconfigured with 72 speech elements (allophones), 43 sound effects, and 12 DTMF Touch Tones. Through
the selection of these MSA components and in combination with the control of the pitch, rate, bend,
and volume parameters, the user has the ability to produce unlimited phrases and sound effects, with
thousands of variations, at any time. This is not recorded waveforms or sound fragments but truly synthetic sound.
The SpeakJet chip has a demo mode in which it plays all available sounds over and over. Please listen:
I used Audition to create a WAV file that contains only the 72 allophones of the demo mode, and to store their
boundaries into the so called "cue list" - this is a data list inside each WAV file which can be used to store the
positions and lengths of sections within the audio recording. It's pretty much the same as the track list of CDs.
Cue list entries can also have named labels and written comments, which, BTW, makes the cue list very useful for
transcribing EVP recordings.
The above WAV file can be opened in EVPmaker as a source audio file. To play the allophones in random order,
the EVPmaker option "Use Cue List" must be selected. "Z-Cross" needs not to be selected since the boundaries of
the allophones in the cue list are already aligned to zero-crossings of the waveform to avoid audible clicks
between the allophones during playback.
To use EVPmaker with SpeakJet Allophones, first download the
SpeakJet allophones file
to your computer. Next, start up EVPmaker, click the "Open" button (1) and select the downloaded SpeakJet
allophones file (2). Make sure that the "Use Cue List" option (3) is selected. Click the button with the
yellow arrow (4) to start the randomized playback of the allophones.
Record the EVPmaker sound via a microphone into a recorder - this may be a digital voice
recorder, a cassette recorder, or even an audio recording software on the same computer. Say or ask something, and
leave pauses between your questions to allow for responses. With some practice you may be able to hear EVP responses
immediately; in this case you could try to to conduct a direct two-way communication. Don't let the session last for
longer than 5 minutes. When you listen to the recording afterwards, you will probably find more EVP. Instead of making
one single session of several minutes with a number of questions, you can also make multiple short recordings, each with
only one question asked, and played back immediately to get the answer. This might me easier than a direct conversation
at the beginning.
If you click the "Rec" button in EVPmaker before you start the randomized playback, the allophones will be
automatically logged into an EVP session file. To analyze an EVP, you can click the "Exp" button to export it into
a new WAV file. If you open such an exported WAV file in a sound editor that supports cue lists such as Audition,
you will see in the cue list the allophones the EVP is made of. (If you don't see the cue list in Audition, select
"View > Show Cue List" from the menu.) If you double-click an individual cue list entry, the corresponding section
within the wave file will be selected; click the Play button to listen to it. This way you can examine exactly of which
speech elements an EVP has been composed of (maybe interesting for a scientific research).
At my first attempt in using the SpeakJet Allophones file in EVPmaker, I heard only babble, but after a few attempts
I suddenly heard a rather clear (at least to me) voice saying:
After I had posted my results in the
EVP-ITC Yahoo Group ,
some of the members there made their own experiments using the EVPmaker and SpeakJet allophones combination, and some of them
got remarkable results with often immediate, meaningful, and relevant EVP responses. Here are some audio examples
of other researchers who used EVPmaker + Allophones:
This is a compilation of multiple short sessions, 20 seconds or so, where Margaret asked a question, got an answer,
then asked the next question based on what she was hearing, and so on. In the following transcript, M indicates Margaret's
questions, and A indicates "Arthur's" responses.
M: "Can you say A B C?"
A: "A B C D E" (this answer is sort of buried in the main sound)
M: "How about Tom Butler?"
A: "Tom Butler."
M: "Do you know the name of either of Tom Butler's cats?"
A: "(Sh)asta's their cat." (the "Sh" sound is weak or missing)
M: "Do you have any pets?"
A: "I have a pet."
M: "I heard you say I have a...?"
M: "You have a what?"
M: "What kind of pet do you have?"
A: "An animal." (LOL - ask a silly question, get a silly answer!)
M: "I understand you have a pet animal, but what is the species?"
A: "A purple deer."
M: "OK, I could have sworn you said you have a purple deer."
A: "I told you purple deer."
M: "OK, I'm gonna close for the evening. Can you tell me goodbye?"
A: "I love you."
M: "I heard you say I love you. Love you too."
A: "Thank you."