I think this might be exactly what we need to start over with a deeper analysis of the audio that extends beyond what we've interpreted as just binary. We do need to perhaps gather more, cleaner audio samples that have some sort of quality control though; how many UAs are there still in the hands of CMDRs? Have any more been found since the beginning of last week? Also, is there any variation in the chitters or are they all identical?
UNKNOWN ARTEFACT: Sound Engineers, apply here!
- Thread starter sohara
- Start date
I think this might be exactly what we need to start over with a deeper analysis of the audio that extends beyond what we've interpreted as just binary. We do need to perhaps gather more, cleaner audio samples that have some sort of quality control though; how many UAs are there still in the hands of CMDRs? Have any more been found since the beginning of last week? Also, is there any variation in the chitters or are they all identical?
I think it's clear the sequences are not binary. In the first full segement of the DigitalScreams's UA_1 sequence there are 4 distinct tones. This has been mentioend a number of times by various people, but always gets overlooked. It is this kind of thing that I hope a shared approach to analysis with a common reference will allow us to discuss a lot more clearly. Again, once I am able I will upload the first of my marked up audacity files (I've only done one so far!)
My belief at the moment is that all the chitters are identical - however it is all open to discussion!
The number of tones, missing 'bits' and the different howls has always suggested to me that there's a large amount of fidelity that we're losing with the current transcription work. If the four tones are base-4 (might make sense to Thargoids as they have four arms apparently) and if the UA community can use your software to extract them more reliably than by ear, then I think we could make significant progress.
Last edited:
How do you plan to distingush the tone if it was inside the howl?
Don't forget that this is a recording in stereo. If the ship or the UA moves a bit during recording, the wave will be slightly different. Plus the left and the right channel will have a visible difference.
I checked the distances between the tones real quick. They always speed up. Each section starts around 4 secs tone frequency and ends at around 3.5-3.6 secs.
This can clearly identifiy the begining and the end of a section.
Also, the chittering is always the same. It has a tone like - - \\__ /- and then a dolphin noise.
The visual cues may worth a note as well:
https://www.youtube.com/watch?v=y2FQzPIZqms
https://www.youtube.com/watch?v=clbLEg3bk1o
The signal pattern is always the same.
The thing lights up and plays the howling or parity signal.
Plays the same chittering signal.
Powers down with the decaying dolphin noise.
Starts the high-low tone sequence.
The object does one full rotation in 3 mins.
The /. .\ pattern is all over it.
It has 7 lines that lit up. The 3rd from the front is different. It has 2 dots, while the others are continuous. Short, Short, Short+dots, Long, Medium, Long, Thich Medium. System pattern, like planets and sizes? Or maybe the count system, like 7 tones in a segment?
Don't forget that this is a recording in stereo. If the ship or the UA moves a bit during recording, the wave will be slightly different. Plus the left and the right channel will have a visible difference.
I checked the distances between the tones real quick. They always speed up. Each section starts around 4 secs tone frequency and ends at around 3.5-3.6 secs.
This can clearly identifiy the begining and the end of a section.
Also, the chittering is always the same. It has a tone like - - \\__ /- and then a dolphin noise.
The visual cues may worth a note as well:
https://www.youtube.com/watch?v=y2FQzPIZqms
https://www.youtube.com/watch?v=clbLEg3bk1o
The signal pattern is always the same.
The thing lights up and plays the howling or parity signal.
Plays the same chittering signal.
Powers down with the decaying dolphin noise.
Starts the high-low tone sequence.
The object does one full rotation in 3 mins.
The /. .\ pattern is all over it.
It has 7 lines that lit up. The 3rd from the front is different. It has 2 dots, while the others are continuous. Short, Short, Short+dots, Long, Medium, Long, Thich Medium. System pattern, like planets and sizes? Or maybe the count system, like 7 tones in a segment?
Last edited:
This appears to be a stereo track and the signals on both channels look pretty similar, almost identical - but looking very closely, there are very subtle differences.
Has anyone tried subtracting one channel from the other - see if there is some sort of code hidden within one channel?
It is pretty easy to remove the howl given a long enough recording. I've talked about how to do this earlier in this thread - at least I think it was this one. Only takes a few mins and I plan to make a quick video demo as I think it would be useful for more people to understand how it is done as this seems the be one of the biggest causes of transcription errors. Will be Monday at the earliest that I can get to this though.
I always mix the two channels down to mono. The stereo effect if determined by the position of the ship that is making the recording, so should not be meaningful. Following mix the channels down, I also normalise the signal to help bring out any details.
It's this kind of thing I'm going to look at in more detail. The more structure we can find, the closer we are to understanding what it is we have! There are some other interesting timing features, such as the timing the howls being regular - except for the timing between the first two where it is quite variable. One thing is clear though - there is a LOT of structure so there is definitely information in there...
Interesting, I've not looked at that. I'm concentrating on the audio at the moment, but it is useful to tie it back with other evidence.
I don't think it's as important as the audio to be honest. Michael Brookes told us to "listen to it" after all.
Phyl, if people are going to capture more audio to use with your tool, do you have any advice on what would make the recordings higher quality and easier to work with?
When a group of CMDRs gathered to look at Red Wizzard's UA, the UA's degradation percentage was different for each of the commanders.
In my instance, the UA didn't degrade at all, it stayed at 100% forever, until we started scooping up duplicates.
Would it be possible to try to recreate this situation in order to make a much longer recording of the UA sound?
In my instance, the UA didn't degrade at all, it stayed at 100% forever, until we started scooping up duplicates.
Would it be possible to try to recreate this situation in order to make a much longer recording of the UA sound?
Hi,
I thought I give it a shot. I used the recordings from He Bo (can't find a reference to these right now but these are more or less the same as all other recordings, I think). Didn't find much but I thought I share what I have found. See: http://www.swedorn.net/UA-Spectrum-hebo6.png
1. These markers increase in amplitude over time. Maybe linked to decay? Otherwise these are too regular to contain any more information.
2. See 1.
3. Interesting! Amplitude of these gets lower over time. Also present higher up in the spectrum and (maybe) a bit lower (just above 4. but not sure). These repeat exactly every 26.006 seconds. Can't tell whether these are just noise (engines?) or actually contain any information.
4. Spottet these right away. Period of 5 seconds. Thought they may be some synchronization signal but after thinking longer about it, this could just be noise as these won't align with any other feature (except for number 7).
5. Seems to be common knowledge: Two types of whale-like sounds. Repeat (about) every 61 seconds.
6. Two types. No more new information.
7. Sharp drop in amplitude every 5 seconds for a wide frequency range. These are aligned with 4.
Since I don't know which sounds are comming from the UA and which sounds are produced by the ship, I propose the following: Someone make two recordings: 1. in front of the UA and 2. in front of some other canister (with the same ship, obviously). Then one can build a noise profile from 2 and filter 1.
Cheers, CMDR Slo
I thought I give it a shot. I used the recordings from He Bo (can't find a reference to these right now but these are more or less the same as all other recordings, I think). Didn't find much but I thought I share what I have found. See: http://www.swedorn.net/UA-Spectrum-hebo6.png
1. These markers increase in amplitude over time. Maybe linked to decay? Otherwise these are too regular to contain any more information.
2. See 1.
3. Interesting! Amplitude of these gets lower over time. Also present higher up in the spectrum and (maybe) a bit lower (just above 4. but not sure). These repeat exactly every 26.006 seconds. Can't tell whether these are just noise (engines?) or actually contain any information.
4. Spottet these right away. Period of 5 seconds. Thought they may be some synchronization signal but after thinking longer about it, this could just be noise as these won't align with any other feature (except for number 7).
5. Seems to be common knowledge: Two types of whale-like sounds. Repeat (about) every 61 seconds.
6. Two types. No more new information.
7. Sharp drop in amplitude every 5 seconds for a wide frequency range. These are aligned with 4.
Since I don't know which sounds are comming from the UA and which sounds are produced by the ship, I propose the following: Someone make two recordings: 1. in front of the UA and 2. in front of some other canister (with the same ship, obviously). Then one can build a noise profile from 2 and filter 1.
Cheers, CMDR Slo
Don't know whether it fits the sensor pulses but I thought the sensors only make a noise if you zoom in on the sensor panel. The pulses are roughly at 1520 Hz (forgot to include a scale). So I don't think these are related but I'll try to capture the sensor pulses from my Asp.
SOme interesting finds Slo. 
I started a thread yesterday about potentially using Audacity labels to mark up interesting features in recordings, and some software I'm writing that can analyse them. If these turn out to be related to the UA and not engine/background noises, then I can add tags for them to my software and we can add them to the mix.
https://forums.frontier.co.uk/showthread.php?t=145917
I started a thread yesterday about potentially using Audacity labels to mark up interesting features in recordings, and some software I'm writing that can analyse them. If these turn out to be related to the UA and not engine/background noises, then I can add tags for them to my software and we can add them to the mix.
https://forums.frontier.co.uk/showthread.php?t=145917
Ok! After unplugging my cheap headset and plugging in the more serious headphones, I discovered the galaxy is a noisy place. Even 20k ly from Sol and 1500 ly above the galactic plane. That is also why I can't change ships right now and compare the ambient noise of my Asp with other ships. Anyhow, the sensor pulses are really always there and repeat exactly every 3 seconds in the frequency band BELOW 1,5kHz. But this might only be the case for my installed sensors and ship type. Sadly, the only thing I can state for sure right now is that if the recording was made with an Asp with similar sensors in a quite place, the pulses must have come from either the UA or any other source of noise in the recording chain.
So, are there any recordings of the UA of which we know which ship was used and are there any recordings of that ship idling in (deep) space? If not, which recording would be best to analyse next?
Phyl said:
Interesting! I'll have a look.
[LONG DISTANCE MESSAGE RECEIVED]
retransmited by an unknown explorer who received that signal from deep space, near Epsilon 69.
[video=youtube;h9jAJsR2NHc]https://www.youtube.com/watch?v=h9jAJsR2NHc&feature=youtu.be[/video]
Summarizing the findings:
Well, I started looking at the sound without even checking this thread xD. I didn´t know it existed. Anyway, what I found:
-The "Clicks" of the Howls (that I called "main signal" in my video) doesn´t seem to follow any pattern that I can distinguish.
-The Purrs (that I called "pre-signal" in my video) seem to follow a clear pattern, sometimes overlapping with the howls.
My theory is that the Purrs are actually a repeated message. An SOS, warning or so. The Howls may be a more detailed message, such as coordinates.
Hope it helps.
retransmited by an unknown explorer who received that signal from deep space, near Epsilon 69.
[video=youtube;h9jAJsR2NHc]https://www.youtube.com/watch?v=h9jAJsR2NHc&feature=youtu.be[/video]
Summarizing the findings:
Well, I started looking at the sound without even checking this thread xD. I didn´t know it existed. Anyway, what I found:
-The "Clicks" of the Howls (that I called "main signal" in my video) doesn´t seem to follow any pattern that I can distinguish.
-The Purrs (that I called "pre-signal" in my video) seem to follow a clear pattern, sometimes overlapping with the howls.
My theory is that the Purrs are actually a repeated message. An SOS, warning or so. The Howls may be a more detailed message, such as coordinates.
Hope it helps.
I posted this in the main thread but I'd like to run this by the audio experts and see what develops. During my latest exploration trip since I started hunting from UAs I went about scanning different celestial bodies but this time with all the music turned OFF. Much to my surprise the UA has very similar sound patterns and some sounds to the ship's detail scan sounds. If anyone can record that and compare it would be very interesting. I can almost hear the same type of chittering sound that the UAs do in the scan sound.
I read that it has been tried already, but I´m thinking about that the high beeps of the Purrs may be some sort of morse? But we need to know how to sepparate the different letters. Anyone has an idea? Anything we have missed?
A very long shoot here but iam reading Michael Brookes Book, Elite legacy.
And in one part of it a character sits in a bar and do not enjoy the new type of music the young people listen to and he explain the music like this:
"A loud din of discordant notes assaulted his ears with anti-instruments. He heard about this music trend on the newsfeeds."
And the interesting part
"Apparently you take a perfect tone for a note and then subtract the same note from the chosen instrument. It create a shadow of the original sound."
And in one part of it a character sits in a bar and do not enjoy the new type of music the young people listen to and he explain the music like this:
"A loud din of discordant notes assaulted his ears with anti-instruments. He heard about this music trend on the newsfeeds."
And the interesting part
"Apparently you take a perfect tone for a note and then subtract the same note from the chosen instrument. It create a shadow of the original sound."
The fundamental problem at the moment is that we haven't found a repeating sequence in these sounds.
I'm still digging around to try and see if any of the characteristics suggest that there is information contained in these sounds. We know that there is some pattern to the sequences; never the same sound three times in a row. Finding anything past that is, so far, proving tricky.
