RTLSDR-Airband VS inaSpeechSegmenter

You're all correct, except there are not that many $30 SDRs out there. Also, most SDRs have clearly superior capabilities to the RTL-SDR, though those capabilities often are not required.

BTW, for simple AM/FM demodulation I cannot recommend enough the handy https://github.com/charlie-foxtrot/RTLSDR-Airband. For example most of LiveATC.net feeds run on it.

rtlsdr_airband, but again requires knowing the frequencies of interest ahead of time.

You may want to give RTLSDR-Airband a look. Its designed for analog AM/NFM voice traffic and can be configured to operate on any frequency your rtlsdr dongle can handle. It can also be configured to output to an mp3 file, in addition to other things like IceCast.

There’s an example service config file in the repo: https://github.com/charlie-foxtrot/RTLSDR-Airband/blob/main/init.d/rtl_airband.service

I have been streaming the output from RTLSDR-Airband (https://github.com/charlie-foxtrot/RTLSDR-Airband) with icecast2 for several (5?) Years now, and I have some time and inclination to look at reducing latency (currently best is 10s, but often much more.)

https://github.com/szpajder/RTLSDR-Airband Might be what you are looking for, although I don't know the performance requirements. It has a scan mode detailed here: https://github.com/szpajder/RTLSDR-Airband/wiki/Configuring-channels-for-scan-mode

I have a little hobby project where I record an FM radio music station using a SDR and then remove all the non-music portions for offline listening. I like the music selections the DJs pick, but I prefer not to listen to the DJ commentary and the advertisements.

I evaluated three methods of recording: analog capture from a standalone FM receiver, using this nrsc5 library to record the "HD" radio stream, and using an AirSpy SDR with this library: https://github.com/jj1bdx/airspy-fmradion

Recording the "HD" (what a misnomer) radio was nice in that there was no hiss or multipath effects, but in comparison to the other methods the digital compression artifacts became impossible to un-hear. It seems to top out at about 96 kbps

The airspy-fmradion library has some nice stuff in it to address multipath, resulting in the best audio quality of the three methods I tested.

I use https://github.com/ina-foss/inaSpeechSegmenter to identify which segments of the recordings are speech vs. music.

One key library that makes it possible is this one: https://github.com/ina-foss/inaSpeechSegmenter

After looking for a few options, I came across inaSpeechSegmenter. It is a speech segmenter and if you pass it an audio file, it returns the time segments of noises and music.