This discussion moved to here: https://quantnmr.com/pposts/Greensleeves_Numbers_Station/
I’m just using this space to dump some files so they can be shared.
Recently, there was a new fake numbers station reading oddly high-pitched numbers between 0 and 25 (probably letters 1-26, so the entire alphabet). The numbers are grouped into 5 number groups (kind of a standard for numbers stations). This occurred during the day (US eastern summer time) on 10 September 2024. I transcribed these numbers to text so I could try to decrypt them, assuming this is a gag and it’s just a simple substitution cipher. My hope is there are enough characters here to apply simple frequency analysis.
The numbers are (with ‘|’ between the five number groups):
19 2 22 25 7 | 4 4 3 19 25 | 22 9 0 17 8 | 19 3 0 1 21 | 7 1 3 5 22 | 14 19 0 5 0 | 25 19 0 0 17 | 0 1 16 10 22 | 19 2 22 25 7 | 4 4 3 19 14 | 2 7 9 22 17 | 3 9 22 15 8 | 1 15 22 25 19 | 2 22 25 8 17 | 22 0 5 19 2 | 22 7 18 22 14 | 13 25 22 7 19 | 8 25 22 14 10 | 2 0 15 22 14 | 19 25 0 23 19 | 2 22 17 7 1 | 15 22 1 14 17 | 7 9 22 0 8 | 25 11 3 1 15 | 7 1 15 13 17 | 7 3 21 15 0 | 21 3 1 3 0 | 1 0 9 22 25 | 7 17 17 17 3 | 5 22 19 2 22 | 23 4 8 3 17 | 15 19 2 22 3 | 25 13 3 19 3 | 22 14 19 25 7 | 18 8 14 7 1 | 15 25 7 9 7 | 16 22 19 2 22 | 22 7 25 19 2 | 5 0 25 16 22 | 19 19 3 1 16 | 19 2 22 4 7 | 17 7 1 13 22 | 0 5 1 7 19 | 8 25 22 4 8 | 19 1 0 21 0 | 25 22 10 22 7 | 25 22 21 7 1 | 23 14 10 3 5 | 19 7 1 15 13 | 17 22 9 22 25 | 10 22 11 1 0 | 10 19 2 22 2 | 3 15 15 22 1 | 18 7 19 2 14 | 7 1 15 19 2 | 22 7 1 13 3 | 22 1 19 10 7 | 23 14 10 22 10 | 3 17 17 14 7 | 4 0 19 7 16 | 22 19 2 22 3 | 25 21 7 13 2 | 3 1 22 14 25 | 22 13 17 7 3 | 21 19 2 22 5 | 3 22 17 15 14 | 7 1 15 5 25 | 22 22 7 17 17 | 13 25 22 7 19 | 8 25 22 14 0 | 18 18 25 22 14 | 14 22 15 4 23 | 19 2 22 3 25 | 16 25 22 22 15 | 10 22 15 0 1 | 0 19 14 22 22 | 11 18 0 10 22 | 25 4 8 19 4 | 7 17 7 1 13 | 22 19 2 22 22 | 7 25 19 2 4 | 22 17 0 1 16 | 14 19 0 19 2 | 0 14 22 10 2 | 0 25 22 14 18 | 22 13 19 3 19 | 19 2 22 25 7 | 4 4 3 19 25 | 22 9 0 17 8 | 19 3 0 1 4 | 22 16 3 1 14 | 1 0 10 19 2 | 22 25 22 3 16 | 1 0 5 19 2 | 22 7 18 22 14 | 3 14 0 9 22 | 25 5 0 25 19 | 2 22 4 8 25 | 25 0 10 14 5 | 0 25 5 25 22 | 22 15 0 21
A histogram of the frequency of the numbers is given below:
So the most popular letter is ’22’, which is most likely ‘e’ or ‘t.’ I’ll update you if I make more progress on this.
The audio files for the recording are given below:
Recording courtesy of Skipmuck (later in the afternoon):
Recording courtesy of ChrisSmol (morning time):
My recording of the morning transmission:
A similar recording courtesy of ChrisSmol from even earlier in the morning – with a different tone of voice and different numbers (yet to be transcribed)
Decoding:
So far I have the following decode key (in Python dictionary form):
key = { 22: 'E',
19: 'T',
2: 'H',
}
Which decodes to:
THE 25 7 4 4 3 T 25 E 9 0 17 8 T 3 0 1 21 7 1 3 5 E 14 T 0 5 0 25 T 0 0 17 0 1 16 10 ETHE 25 7 4 4 3 T 14 H 7 9 E 17 3 9 E 15 8 1 15 E 25 THE 25 8 17 E 0 5 THE 7 18 E 14 13 25 E 7 T 8 25 E 14 10 H 0 15 E 14 T 25 0 23 THE 17 7 1 15 E 1 14 17 7 9 E 0 8 25 11 3 1 15 7 1 15 13 17 7 3 21 15 0 21 3 1 3 0 1 0 9 E 25 7 17 17 17 3 5 ETHE 23 4 8 3 17 15 THE 3 25 13 3 T 3 E 14 T 25 7 18 8 14 7 1 15 25 7 9 7 16 ETHEE 7 25 TH 5 0 25 16 ETT 3 1 16 THE 4 7 17 7 1 13 E 0 5 1 7 T 8 25 E 4 8 T 1 0 21 0 25 E 10 E 7 25 E 21 7 1 23 14 10 3 5 T 7 1 15 13 17 E 9 E 25 10 E 11 1 0 10 THEH 3 15 15 E 1 18 7 TH 14 7 1 15 THE 7 1 13 3 E 1 T 10 7 23 14 10 E 10 3 17 17 14 7 4 0 T 7 16 ETHE 3 25 21 7 13 H 3 1 E 14 25 E 13 17 7 3 21 THE 5 3 E 17 15 14 7 1 15 5 25 EE 7 17 17 13 25 E 7 T 8 25 E 14 0 18 18 25 E 14 14 E 15 4 23 THE 3 25 16 25 EE 15 10 E 15 0 1 0 T 14 EE 11 18 0 10 E 25 4 8 T 4 7 17 7 1 13 ETHEE 7 25 TH 4 E 17 0 1 16 14 T 0 TH 0 14 E 10 H 0 25 E 14 18 E 13 T 3 TTHE 25 7 4 4 3 T 25 E 9 0 17 8 T 3 0 1 4 E 16 3 1 14 1 0 10 THE 25 E 3 16 1 0 5 THE 7 18 E 14 3 14 0 9 E 25 5 0 25 THE 4 8 25 25 0 10 14 5 0 25 5 25 EE 15 0 21
