Adjusting mp3 file output sample rate and dynamic range

[ejackson]

Hello,

I would like to improve the quality of the .mp3 file outputs from airband. Currently the sample rate is 8kHz and the output volume is is really low. I cannot find the settings to adjust these, and any help will be appreciated.

Thank you,

Edmund

---

My config file looks like this:

File outputting listener based on RTLSDR.

devices:
({
type = "rtlsdr";
index = 0;
gain = 40;
centerfreq = 462.500;
afc = 1;
channels:
(
{
freq = 462.5625;
modulation = "nfm";
squelch_threshold = -15;
outputs: (
{
type = "file";
directory = ".";
#include_freq = true;
filename_template = "1";
split_on_transmission = true;
}
);
},

... etc

[charlie-foxtrot]

If you compile in NFM support the output audio will be 16 kHz, but this may not actually improve the audio quality any.

Doing a higher audio bitrate will only preserve audio quality. The AM demodulator runs at 8 kHz and the NFM demodulator runs at 16 kHz.

As for turning up the volume, that should be a straight forward code change, I'll add that to the backlog

[ejackson]

Thank you.

16k vs 8k: I beileve I compiled on unstable, with NFM enabled and still only getting 8k. I've inlined a log below. Is this a bug ?

Volume: if its a quick-fix I'd be happy to take a crack at it.

Thanks,

Edmund

---

$ git checkout unstable
branch 'unstable' set up to track 'origin/unstable'.
Switched to a new branch 'unstable'

$ git pull
Already up to date.

$ cmake ../
-- Build type not specified: defaulting to Release
-- Found lame includes: /usr/include/lame/lame.h
-- Found lame library: /usr/lib/x86_64-linux-gnu/libmp3lame.so
-- Checking for SoapySDR
-- SoapySDR found, /usr/include, SoapySDR
-- RTLSDR-Airband configuration summary:

-- - Build type: Release
-- - Operating system: Linux
-- - SDR drivers:
-- - librtlsdr: requested: ON, enabled: TRUE
-- - mirisdr: requested: OFF, enabled: FALSE
-- - soapysdr: requested: ON, enabled: TRUE
-- - Other options:
-- - Platform: default
-- - Broadcom VideoCore GPU: FALSE
-- - NFM support: ON
-- - PulseAudio: requested: ON, enabled: TRUE
-- - Profiling: requested: OFF, enabled: FALSE
-- - Icecast TLS support: TRUE
-- Configuring done
-- Generating done
-- Build files have been written to: /home/edmund/Code/RTLSDR-Airband

$ cd..
$ make
$ sudo make install

$ rtl_airband -v
RTLSDR-Airband version 4.0.3-3ac8ef4

$ rtl_airband -f -c

$ mplayer 4_20230303_171831.mp3
MPlayer 1.4 (Debian), built with gcc-12 (C) 2000-2019 MPlayer Team
do_connect: could not connect to socket
connect: No such file or directory
Failed to open LIRC support. You will not be able to use your remote control.

Playing 4_20230303_171831.mp3.
libavformat version 59.27.100 (external)
Audio only file format detected.
Load subtitles in ./

Opening audio decoder: [mpg123] MPEG 1.0/2.0/2.5 layers I, II, III
AUDIO: 8000 Hz, 2 ch, s16le, 8.0 kbit/3.12% (ratio: 1000->32000)
Selected audio codec: [mpg123] afm: mpg123 (MPEG 1.0/2.0/2.5 layers I, II, III)

AO: [pulse] 8000Hz 2ch s16le (2 bytes per sample)
Video: no video
Starting playback...
A: 2.2 (02.1) of 3.0 (03.0) 0.1%

---

With this config file

---

devices:
({
type = "rtlsdr";
index = 0;
gain = 40;
centerfreq = 462.500;
afc = 1;
channels:
(
{
freq = 462.5625;
modulation = "nfm";
squelch_threshold = -15;
outputs: (
{
type = "file";
directory = ".";
#include_freq = true;
filename_template = "1";
split_on_transmission = true;
}
);
},
{
freq = 462.5875;
modulation = "nfm";
squelch_threshold = -15;
outputs: (
{
type = "file";
directory = ".";
#include_freq = true;
filename_template = "2";
split_on_transmission = true;
}
);
},
{
freq = 462.6125;
modulation = "nfm";
squelch_threshold = -15;
outputs: (
{
type = "file";
directory = ".";
#include_freq = true;
filename_template = "3";
split_on_transmission = true;
}
);
},
{
freq = 462.6375;
modulation = "nfm";
squelch_threshold = -15;
outputs: (
{
type = "file";
directory = ".";
#include_freq = true;
filename_template = "4";
split_on_transmission = true;
}
);
},
{
freq = 462.6625;
modulation = "nfm";
squelch_threshold = -15;
outputs: (
{
type = "file";
directory = ".";
#include_freq = true;
filename_template = "5";
split_on_transmission = true;
}
);
},
{
freq = 462.5875;
modulation = "nfm";
squelch_threshold = -15;
outputs: (
{
type = "file";
directory = ".";
#include_freq = true;
filename_template = "6";
split_on_transmission = true;
}
);
},
{
freq = 462.7125;
modulation = "nfm";
squelch_threshold = -15;
outputs: (
{
type = "file";
directory = ".";
#include_freq = true;
filename_template = "7";
split_on_transmission = true;
}
);
}
);
}
);

[charlie-foxtrot]

I haven't had a chance to look at the code closer on the 8k vs 16k but will get back to you on that.

For opening a PR on the volume change, that would be great (please fork and open a PR to the unstable branch). I suspect most of the code change will be on supporting the config parameter. A float defaulting to 1.0 that is just multiplied to the WAV after demodulation here-ish should be all it takes, but the config file should support setting it per-channel (pre-mixer) and support multiple values for "scan-mode". Take a look at notch as an example.