layla
New Member
Posts: 42 
|
Post by layla on Apr 22, 2024 15:42:05 GMT
thanks, i did share the program intention repeater wav with others my friends on discord that want as me to become in a natural way females and even for who want to become male, i did share the audio intentions i made with my friends. again sorry for my bad english.
|
|
|
|
Post by reden on Apr 22, 2024 15:44:30 GMT
Minimum Frequency (Hz) 100 Maximum Frequency (Hz) 5000 You'll have to change the maximum frequency to like 25000 or something. It will see if it's in that range.
You can look between 20000 and 25000. Frequencies in that range won't probably play in speakers.
The only certainty is that speakers can play frequencies up to 20000 Hz. |
|
|
|
Post by AnthroHeart on Apr 22, 2024 15:46:15 GMT
Most algorithms aren't lossless. Use FFMpeg to speed up losslessly. There's a way. This worked for me to shift from 440Hz to 17000Hz frequency, when the sampling rate is 44100.
ffmpeg -i "audio.mp3" -af "asetrate=44100*17000/440,aresample=44100,rubberband=tempo=440/17000" -acodec flac -compression_level 12 output.flac
ChatGPT 4 says:
In summary, while 440 Hz can be used as a starting point, considering a lower base frequency, perhaps closer to the average fundamental frequencies of human speech (around 85-255 Hz), might be more effective. This lower starting point could facilitate a more accurate and intelligible representation of the human voice in the ultrasonic range. The exact choice would also depend on the specific application and the conversion method used.
|
|
layla
New Member
Posts: 42
|
Post by layla on Apr 22, 2024 15:47:06 GMT
oh i was forget of say that... i did found on this forum a program that create automatically a subliminal with morphic field but just for linux, do exist a windows version ?
|
|
layla
New Member
Posts: 42
|
Post by layla on Apr 22, 2024 15:51:30 GMT
in public small chat : sound: New tool to create automatically a subliminal with morphic field on Ubuntu-based systems: github.com/soundenergies/auto-morphic-subliminal Jul 21, 2022 at 8:35pm links do not works
|
|
|
|
Post by reden on Apr 22, 2024 16:58:36 GMT
This worked for me to shift from 440Hz to 17000Hz frequency, when the sampling rate is 44100.
ffmpeg -i "audio.mp3" -af "asetrate=44100*17000/440,aresample=44100,rubberband=tempo=440/17000" -acodec flac -compression_level 12 output.flac
ChatGPT 4 says:
In summary, while 440 Hz can be used as a starting point, considering a lower base frequency, perhaps closer to the average fundamental frequencies of human speech (around 85-255 Hz), might be more effective. This lower starting point could facilitate a more accurate and intelligible representation of the human voice in the ultrasonic range. The exact choice would also depend on the specific application and the conversion method used.
I don't think this matters at all |
|
|
|
Post by AnthroHeart on Apr 22, 2024 17:09:48 GMT
ChatGPT 4 says:
In summary, while 440 Hz can be used as a starting point, considering a lower base frequency, perhaps closer to the average fundamental frequencies of human speech (around 85-255 Hz), might be more effective. This lower starting point could facilitate a more accurate and intelligible representation of the human voice in the ultrasonic range. The exact choice would also depend on the specific application and the conversion method used.
I don't think this matters at all If you're shifting up frequency by 17kHz so that 440Hz becomes 17440Hz, wouldn't already existing 15kHz frequencies be over 30kHz? That's beyond the ability of speakers.
Would most of human speech still be able to be reproduced by speakers in ultrasonic?
ChatGPT says:
Overall Frequency Range: Combining both speech and singing, the frequency range typically spans from as low as 80 Hz to as high as 5,000 Hz or more when considering harmonics and formants. This range covers the fundamental frequencies and the important harmonics and formants that contribute to voice clarity and color.
|
|
|
|
Post by sound on Apr 22, 2024 18:30:35 GMT
Anything above 20kHz sampled at 44.1kHz will alias, also, since it's half the Nyquist frequency
|
|
|
|
Post by reden on Apr 22, 2024 18:56:14 GMT
Most algorithms aren't lossless. Use FFMpeg to speed up losslessly. There's a way. This worked for me to shift from 440Hz to 17000Hz frequency, when the sampling rate is 44100.
ffmpeg -i "audio.mp3" -af "asetrate=44100*17000/440,aresample=44100,rubberband=tempo=440/17000" -acodec flac -compression_level 12 output.flac
Without rubberband the audio length decreases a lot, so it's excellent to pack a lot of affirmations in a small file. |
|
