The first thing that strikes people when they hear the name ‘ghost’ is the sound of laughter.
But a team of scientists from the US and UK have found that laughter is a useful metaphor for the phenomenon.
The study, published in Nature Communications, analysed the sounds produced by more than 30,000 different kinds of sounds to investigate how they are perceived.
The researchers believe this could lead to new technologies that can be used to create artificial intelligences or even create new kinds of sound.
In their paper, co-author John A. Kaczmarek and colleagues looked at the acoustic properties of sound to understand how it can be perceived.
They used a series of microphones to measure the frequency and amplitude of the sound waves, and analysed the spectra of the sounds they recorded.
When they compared the spectre of the recordings to the spectres of human laughter, they found that it differed in that laughter sounded more like an acoustic wave, but the sound emitted from the mouth was less.
‘Ghost’ could be used for audio recording, says co-lead author John Kaczan, a doctoral candidate in electrical engineering at the University of California, Berkeley.
The authors also looked at whether the spectrums of the recorded sounds reflected the spectral characteristics of a sound source or whether the acoustic waves emitted from a sound device reflect the spectrum of the source.
This suggests that there might be some way to detect a source of sound by analysing its spectra.
This is not a new idea; previous work by the researchers suggests that ghost sounds are generated by an invisible source that emits an acoustic signal that can then be recorded using a microphone.
But this is the first time that this kind of research has been done in real-world situations.
‘It’s kind of amazing that we could do this in a lab, so this is exciting to us,’ Kaczo said.
‘What we are really excited about is how this can be applied to audio and to other kinds of audio processing.
It’s the next step in the evolution of this technology, so it’s a really exciting time for the field.’
Kacza says that it’s possible that future experiments will be able to use this new knowledge to create ‘sonic ghosts’ or to make audio recordings using sound produced by ‘natural sounds’.
But in this paper, the researchers suggest that this is not the only use for this kind and variety of sounds.
‘There is a lot of interesting applications for this technology that we don’t even know about yet,’ Kacszmareks said.
The scientists have been working with the acoustic and spectroscopic properties of a variety of human noises, including laughter, to try to understand the origin of this phenomenon.
‘If we can capture that same sound in a laboratory, that sound can be recorded as a real acoustic wave and that wave can then go to the next level, like an echo that will be emitted in the next room and then you can record it from there, say in a recording studio,’ KACZANES said.
To understand how ghosts work, Kaczi said that the team will focus on three basic types of sounds: ‘sphonic’, which is the acoustic wave that emits a sound; ‘acoustic phonons’, which are the sound vibrations produced by human mouths and bodies; and ‘auditory phonons’.
They are based on the way in which the human brain generates these sounds, and their relationship to other sounds that are produced by the body.
‘We’ve known that this sort of acoustic phonons are produced in the mouth when there’s a lot that’s going on in the body, like when the person is making a noise, and that’s a type of sound that we can use to record that sound,’ Kanczmareck said.
He added that this study could provide an important step towards this goal, as well as finding ways to record sound from other sources without actually using the human body as a sound processing device.