[Feature Request] Allow lower frequencies in the 2D spectrogram to detect infrasound

[danielmmmm]

It would be very interesting to lower the min. frequency below 20 Hz to detect infrasound. Maybe the FFT size has to be increased for that as well, to increase the resolution?

[tlecomte]

Measurement microphones usually have a flat response only from 20 Hz. For example: https://www.behringer.com/product.html?modelCode=0506-AAA

Of course, that does not mean that infrasound is not interesting, but just not for sound that can perceived by the ear.

What use case do you have in mind @danielmmmm ?

[danielmmmm]

My use case was to investigate whether the humming that plagues me is in my head or from an outside source like a manufacturing plant or a motorway. By now, I am fairly certain that I have some low frequency tinnitus.

That said, I still find the idea of detecting infrasound highly fascinating, just like ultrasound, thermal infrared, or gamma radiation. With a cheap MAX9814 microphone amplifier and electret microphone, I picked up signals, e.g. from cars driving by the house, going down to 20 Hz. So I am certain it's really sound, and not a noise artefact from the amplifier or microphone. And I am planning to make a simple microphone array to increase the sensitivity and maybe pick up infrasound from the nearby chipboard manufacturer. I don't believe that I can hear the chipboard manufacturer at my house. But you can easily hear and feel a very low frequency sound close by, and it would be cool to visualise that.

This would be an experimental feature to detect infrasound more than to quantify it. I also don't know what lower frequency would make sense for a cutoff. My guess is something between 10 and 15 Hz. Although a software cutoff at 1 Hz could make a signal at 15 Hz more visible since the 2D spectrogram seems to render with interpolation and a signal at precisely 15 Hz would probably look like a wide blob around 15 Hz.
Since I am not a programmer, I don't know if adding infrasound detection would be as easy as changing some numbers in the code to reduce the lower frequency limit and increase the FFT resolution, or if it would be a major undertaking involving mathematics and algorithms or fighting Windows drivers and the Windows audio system.

Hm ... since I mentioned ultrasound ... if you find the idea of infrasound interesting, would ultrasound be of interest too? That could visualise animals like bats.