Hello everyone,
I have read your topic “Silent Subliminals [solved?]” a few days ago. And the Edgar’s program works well, because I was able to create a record with the characteristics Lowerey.
And I have two questions:
Thank you.
Try it and see if you get an error. I think it should work.
See this topic: https://forum.audacityteam.org/t/silent-subliminal-function/18148/9
Thank you Steve.
Effectively, the transition to stereo works!
I will look on the 2nd point with your link.
Hello,
Why the frequency range generated during the transition to the Silent Subliminalx is truncated in mp3 format?
Frequencies before conversion (17200 to 21900 Hz) and after conversion (17200 to 20120 Hz).
Thank you.
MP3 achieves small file size by discarding “unimportant” information and squashing the remaining data tightly together. Because very high frequencies are virtually inaudible, that is some of the first information to be discarded.
Hello Steve,
About plug-ins Subliminal.ny, I would be able to limit the high frequency 20,000 Hz.
Which parameters change in the plug-ins Subliminal.ny for example, to have a signal between 14500 Hz and 20000 Hz?
Thank you.
Sorry, I don’t understand your question.
My mp3 player is limited to 20,000 Hz, so how do you generate from an mp3 stereo audio,
a subliminal message with the highest frequency does not exceed 20000 Hz :
I get exceeds 21,000 Hz :
Thanks.
MP3 format is unsuitable for this type of recording. I wrote an explanation of why that is previously in this thread.
thank you steve, then I will have to change the music player because it only plays the mp3!
please do you can write the complete process of edgar subliminal plug in , how not exeed 20000 hertz , because my record with subliminal effect reach " 21000 hz"
at 44,100 frequency , at 14500 hz in subliminal effect and reach 22000 with sublimal effect at 17500 hz …
the code for nyquist is not good , because the ( spoken words ) affirmations are reduced, decreased sound quality !!!
please help me !!!
in france there is a website " subliminal silencieux " , with a subliminal plug in ( 50 euros ) what is the difference with the edgar plug in ( or subliminal ny )
which doesnt deforme the voice …
THANK FOR YOUR ANSWER 
The code is good.
The sound quality is poor because for a sample rate of 44100 Hz and a carrier frequency of 17500 Hz, the audio bandwidth is less than 4500 Hz, which is not enough for high quality audio. You will need to use a lower carrier frequency, or a higher sample rate, or a different method for making the voice “silent”.
thanks steve , ok if i create a project at 14500 hz with subliminal effect and i change carrier frequency k at 96000 it will be ok ???
i think it s necessary to write a GOOD TRUE tutoriel … PLEASE
for the student in subliminal process with audacity …
How do I eliminate a range of frequencies? For example, I would like to eliminate frequencies of 0 to 15Khz from a mp3 file which reportedly
has silent subliminals. This would clean up any environmental sounds and leave just the expected silent subliminal. It is not clear to me how
this can be done with any of the tools within Audacity or better yet how to code this in a hopefully simple Nyquist script which could be added
as a command. Thanks in advance for your help. 
You can eliminate/crop range of frequencies easily with a 30 day trial version of Adobe Audition. It shows the waveform of each frequency in varying shades of color indicating amplitude. I’ve been using this in an attempt to decode the ultrasonics being BOMBARDED into my room at extreme levels. They’re definitely using a piezoelectric transducer…
I actually have another off topic question. Has anyone been able to find a microphone that successfully records up to 20 KHz?
MP3 is unsuitable for AM encoded “silent subliminals” because it does not have good enough high frequency response.
To eliminate (or at least “reduce greatly”) frequencies below 15 kHz, use the “High Pass Filter”, set to 15000 Hz and 48 dB per octave. Audacity Manual
Considering that this is the Audacity forum, I expect most users here will want to do the job in Audacity.
Steve,
Thanks for your response. I have tried using the high pass filter with the maximum db rolloff but it does not delete ALL of the lower
frequencies as you can visually see in the waveform setting when set to spectrogram. In fact, when I use the shifting function subliminal.ny
I also see residue in the lower frequencies which was not moved suggesting a possible installation problem with Audacity on this PC. I have
re-installed Audacity 2.0.6 but still see the “residue” issue. The unsubliminal command also appears to leave a copy of the wave at the
higher frequencies but does move or copy? the audio into a correct range. Using 320bit mp3s seems to address most of the elimination of the higher frequencies of the mp3 transform as I can hear a very legible version of the encoded message as well. I source the messages in wave file format and only the end
product is in mp3 format. I can always use FLAC format as my mp3 player supports this lossless codec in any case for the final product. The Nero
wave editor has a filter toolbox which seems to work like I described in removal of the lower frequencies. Is there a run-time library which I may need
to re-install, delete, or update for Audacity? I would like to do all of the work in Audacity if at all possible.
Question for stocks: What are you actually trying to do?
Filters are never 100%. There is no such thing in audio as a dead cut-off at a particular frequency, though some filters can get quite close.
48 dB per octave is about 250 times quieter one octave below the corner frequency, which is quite substantial. If you require more attenuation than that, there are ways to do it, but there are also likely to be side effects, so “what the job is” is important.
Steve,
I have two goals:
I would like to all of this within Audacity without the need for outside software except for
creating the plain text (subliminal message) which will be created with other audio software
designed for creating such messages easily with built-in (auto panning, random order of message
generation, the ability to add isochronic tones, binaurals, …, etc).
Goal 2 is where I was wanting to be able to easily eliminate the entire range of frequencies.
This is primarily done to eliminate any mask so that when the UHF message is moved into
the audible spectrum, then I would only hear signal degradation based upon the MP3 algorithm
loss, too low of sampling, etc. Since it is not always easy to see the exact range of frequencies
that would need to be eliminated, I was hoping to create a Nyquist script to aid in doing this
functionality. It appears from examining the subliminal.ny script that a modification to the cut
functions might be able to do the job. I also considered using the vocal remover utility but it was
unclear if I could keep the vocals and throw away the remainders (just the opposite of what the karaoke
function) was developed to accomplish. This would help isolate the voice in some masked subliminals
where the message is overpowered by higher decibel sounds. This is a lower priority item on my list for now.
Goal 3 is seeing how to take a silent subliminal created as a low frequency to see how I can recover the message
that was created in that methodology. At least that would show the message is actually recoverable; not that is
can be decoded by the brain.
Goal 4 is to combine these 3 different approaches as the targeted message so you get the actual audible message, the UHF message, and the VLF
message and perhaps some sort of BWE could be added if it does interfere with the other frequency ranges.
Hopefully, I can attach a 20 second sample for eliminating sinus congestion in FLAC format. Maximum file size is too big, sorry.
As far as my installation comments, I am not seeing unsubliminal function actually moving the higher frequencies into the audible spectrum.
It appears to be doing a copy function. That is what led me to believe that I may have an installation problem.
I presume that you have found and read the various topics on this forum about that?
First thing is, do you know how they have been encoded?
Lowrey’s (amplitude modulation) technique is one of the most common, but you need to know how it has been encoded in order to be able to decode it (which makes me wonder if the subconscious brain really can do that all on its own).
Assuming that they are AM encoded, the upper limit for 320 kbps MP3 is 20000 Hz. For the message to be “silent”, the carrier frequency must be at least 15000 Hz, so at best the audio band can be no more than 5000 Hz (just about enough for voice). So that means that you can do the filtering in two stages and get virtually complete isolation.
First, apply two passes of a 48 dB per octave high pass filter at about 14000 Hz. That will give you about 90 dB attenuation at 7000 Hz.
Then decode the “message”. Decoding will bring the message frequency down below 5000 Hz.
Then apply two passes of a low pass filter at about 6000 Hz. That will remove any remaining high frequency sound, leaving just the decoded message.