The Day After Tomorrow

Thanks for the very helpful feedback. Much appreciated. 

I find it challenging to make such a long song without appealing variety

But I soon tired of the constant barrage of rapid fire notes.

That's exactly what I thought too. But sometimes you still need to hear it from others to be absolutely sure. The piece is still too long and the main pattern (the constant barrage of rapid fire notes) is annoying after a while. The issue with the annoying pattern will certainly improve once the vocals enter the scene, because then it will be pushed further into the background. Currently the track is more like the Karaoke version of a song. But I guess it's true that the piece is still too long. Although I have already shortened it from over 5 minutes to about 4:35. Nevertheless, it is too long.

Another problem with the current structure is that the singer will only appear after more than 30 seconds. That is much too late. I probably have to sacrifice the intro then, or at least shorten it drastically. This means that the vibraphone or xylophone-like intro pattern (at 0:15) must also be dropped. That was derived from a motif from the above "day-after-tomorrow" arrangement (I think it was the flute motif that is being played around 0:59). Being played with that xylophone-like sound (actually it is an Absynth patch) it reminded me to the intro pattern of the song "Somebody that I used to know" and I liked that. But in the end it is dispensible.

The reason why the "rapid firing notes" pattern plays throughout the song is because it played a center role during the song development. In the original "day-after-tomorrow" arrangement it is the piano / strings staccato pattern that appears around 0:39. I made a sketch out of the entire container in which this appears and used it to develop the song structure and progression. I really liked it at first.  But it's true, after the five hundredth playthrough you can get a bit sick of it.

I'll think about a potential theme for the lyrics.

Thank you. My current idea are not fictional lyrics, but they tell a story. If it turns out that this doesn't work, I would need alternative lyrics. Or I'll abstract them a bit, we'll see. But I will definitely try to include one line: "The day after tomorrow" :)

Oh and I love these beautiful shoulders and the milky way!

Well, the blonde girl is not there by chance. Neither are the stars.

If the Soundcloud account name is any hint at your intentions, this might be a great basis for a collaboration where someone else may want do a remix or something.

I did set up that account several years ago for exact that purpose. Everyone is welcome to upload something there. Just send me a message via PM and I will send you the password.

Some of that hypnotic repetition will probably be gone when you add vocals and carve out a song structure. 

What I actually have in mind is to fit the vocal melody rhythmically in such a way that it complements the instrumental patterns. This means that the instrumental patterns should not simply be buried by the vocals. If that were to happen, there would probably be a mush of sound and I would have to take something out again. That's also the reason why I put relatively little instrumental movement into the verse sections at the beginning, so that I can also fit a few more words in there. 

The first chorus is still a bit thinned out instrumentally, so a few song lines should fit in there too. The second chorus, when the lead synth plays all the way through, is supposed to be purely instrumental as it is now. 

In the third run-through of the chorus, with full instrumentation and all the effects on top (it's all Arturia's new Efx Motions effect, isn't it cool?), I would actually also like to squeeze in a few vocal lines. That might be a bit difficult, we'll see. If I do so, then only a few lines, probably alternating with the instrumental melody.

This may all sound incredibly difficult, but it's actually not. When I listen to the track, I already have almost the entire melody in my mind's ear anyway. I just have to enter it, try it out to see if it works and then adjust it. 

And with Synfire you can easily move everything back and forth and up and down without having to worry about the harmonies, because it always fits in that respect anyway.

I have attached the template arrangement that I am currently using to create the melody. It contains containers corresponding to the song structure (shifted forward by one bar to make it easier to synchronize with the audio file), the progression and an instrument with a flute sound in which I now enter the melody. If anyone would like to try their hand at a melody too, feel free to do so.

I'm making progress, by the way. What is causing me some trouble is the word "tomorrow". Whoever invented this word obviously didn't think that someone might want to sing it one day. I'm currently listening to a few songs that contain it. I'll probably try to do it like in Avril Lavigne's "Tomorrow" song.

The harmony in your track makes a big difference. It adds an "epic" feel to the hypnotic sequences. 

The progression is actually quite simple. It was developed by "palette surfing" with a Sketch from your "Stac. 2" container of the original "day after tomorrow" arrangement.

The progression noodles around on the G minor chord for the first 35 seconds. Then there are a few chord changes. For the chorus it then changes to the relative key Bb major. You can see it in the enclosed file. 

The "epic" style towards the end is indeed intentional. I hope to align the lyrics with this

I can report some progress here. I now have the rough version of a melody (still with flute sound) and lyrics. What do you think, does it work?  

If anyone has ideas on what can be optimized, please go ahead, the file with the melody and the lyrics (in the comment field of the containers) is enclosed.

Regarding the structure and length of the song, I've decided to finish it as it is for now. Then I'll see if it needs to be shortened (but it probably needs to be).

My next step now will be to transfer this to Synthesizer V.

Regarding a theme for the lyrics, I thought of young post-humans in the year 2424 celebrating the extinction of homo sapiens, an inferior species...

Man, you have ideas, incredible! But I like it. That would definitely emphasize the “Trance” character of the piece you're hearing in it. But how would that translate to the actual vocals? Probably shorter and faster notes, almost Rap, or maybe even actually a Rap section, yes that could work (did you know that Solaria now has a Rap mode; would be a good occasion to try that out). But short and fast notes means many words. Words that first would have to be written. And I would probably use a different kind of instrumentation for this, harder, more shrill. My current approach for the vocals emphasizes more the “Epic” part of the song.

But yes, I like your idea. I will handle it like you do with feature suggestions: “I add it to the list”. Maybe I come back to it with one of the next songs. Somewhere on this forum there is a thread, "How to overcome the writers block”. I have the solution: Just ask for some advice on this forum. Or take one of the arrangements lying around here and do something with it :-)

There are some extreme pitch leaps in it. But if Solaria can handle it, why not. Wide pitch leaps are however uncommon for vocals. 

I have to disagree with that. Pitch jumps are one of the main elements to make a song to sound interesting. A good singer can handle this. Now we will see, how good Solaria actually is. If she can’t do it, she’s fired and I want my money back from Dreamtonics :)

You might need to split notes a lot depending on the lyrics. There will be many more syllables for sure.

No, I have developed the melody hand in hand with the lyrics. I can’t do it any other way. So, the number of notes in the melody is exactly the number of syllables of the lyrics. You can read the lyrics in the comments boxes of the section containers (also visible in the video). If you do so, you will notice that I even managed to squeeze your "the day after tomorrow" phrase three times into the song. That was a fiddle job, I can tell you. But I love challenges like these. There is always a solution.

By the way, about the lyrics: Since English is not my native language, it would be good if someone could check the correctness of the wording, expression, etc. (I could ask others, but then I'd probably get more questions than answers). For instance in Verse 2: What is better "Way to late" or "Far to late"? ("way" would be better in terms of singing)

Next stop: Solaria version.

I now have a raw Solaria version:

I have to say, I'm pretty happy with the result. To me It almost sounds as if Solaria feels what the song is all about.

As expected, pitch jumps are no problem. Rather, the pitch range of the voice is perhaps a little limited. I had to do a bit of trickery here and there. The pitch of the word "sound" at 2:05 was too high and sounded too unnatural. I then added a second voice an octave lower and now it works, I think. At another place I simply changed the pitch of a few notes. Otherwise, however, everything is exactly the same as in the flute melody above.

There's still a lot of work to do on the song, but before I finalize it, I'll probably leave it for a few weeks to get enough distance. 

Thanks for all the comments. Very much appriciated and helpful.

It's emotive, and that's what music is all about, nicht wahr?

Yes, emotions are definitely involved here. The thing about the lyrics (I will add them to the soundcloud page) is that they're not just some imaginary stuff, they're basically telling the story of a real tragedy (in a way that is as abstract as possible of course). That's why I wasn't and still am not sure whether it's really a good idea to do this, especially with this type of music. That's also the reason why I ask so many questions. Either I do it right, or I don't do it at all. If I'm satisfied with the final result of the song, I can write something about the background if anyone cares. 

I'd probably shorten the intro

Yes, I've already decided that the intro part from 0:10 to 0:30 has to go. That solves several problems at once: Firstly, the song becomes shorter, which I think is a good thing. Secondly, the vocals come in much earlier, which is even more important. And thirdly, I get rid of the slightly annoying ringing. I mean, who wants to listen to a ringtone for 30 seconds?

But I will keep the pattern at the very beginning, because it has a certain meaning. It should create a subtle reference to the intro of another song. There are a few other such references in the song.

The vocal interval ~3:28 works, but the same thing 15 seconds later is a bit much. Like too much spice in the soup. You need some time to clear the palette before you do that again.

That's exactly the kind of comment I need. Yes, you're right, now I hear that too. That's where your decades of songwriting experience come in :-) 

I'll try to soften the melody line a bit. I have to go back to it anyway, because I've now noticed that there's also a mistake in the lyrics (in the line "where you are now" Solaria sings the words "you" and "are" the wrong way round, which makes the sentence sound like a question, which is not intended).

Btw, I used to be "Hangdog Cat"

I know. You don't have to work at the secret service to figure that out. .-)

you remember Enigma?

Yes, of course. But I actually had songs from someone else in mind when I was doing this. But with Michael Cretu, we have at least a connection to the right country.

If you had a few drops here and there where only the voice and bass shine through for two bars or so, that would greatly add more depth to the arrangement.

Yes, I understand what you mean. However, it's a bit difficult to implement in the current project phase.The right place where this is perhaps still possible without too much effort is the transition to the instrumental part. I will certainly try it.

But not only the song need drops and brakes. I need a brake from that song too. :)

I miss the milky way! The first picture was so much better ;-)

You can't have everything. These AI tools can be pretty stubborn, you know? :-)

Repeating this vocal level a few times in the song with different lyrics repeating the subject matter.

That would make the song even longer, but it's already too long. Maybe I'll do a remix one day.

Perhaps also the speed of singing could be higher in order to be more in sync with the music ?

Can you tell me where exactly it is out of sync? Of course I can move the notes around without any problems, but I thought it would work where they are now.

Where the notes start a little later, this is probably intentional. This should provide the necessary portion of "human feel". I actually paid attention to every millisecond of the note positions.

Singing and the song at the same tempo for a song section and alternating with a different tempo 

Ah, ok, now I understand. Sounds like a lot of work. But I think I should slowly come to an end with this song. 

There's still enough to do anyway. Shorten the intro, optimizing the melody, rework the instrumentation and thin it out at some places if possible (as Andre suggested), then I want to add a few ad-libs or backing vocals here and there (which first need to be prepared), then the whole thing needs a new mix and then mastering. 

Like the tropical sphere of the song .

I think I have now also understood this comment of yours. You probably mean the lead synth with the strong portamento effect. Interesting. Yes, you can see it that way. Sounds like a deep sip from a Bacardi bottle, doesn't it?

That's a good thought anyway. I'm going to put the song aside for now and let it mature like a good bottle of wine. Or like a bottle of Bacardi in this case. At some point, I'll open the bottle and see what kind of bouquet flows towards me. Then I'll see if there's any point in continuing with it. Or whether I need to lock the song up in the poison cabinet.

After a certain break I am now continuing with this song. The first time I listened to it again, I thought "Ok, really not so bad". So I think it's worth to continue.

I will implement all your suggestions for changes. The intro has already been shortened and reworked. However, with this, the last perceptible connection to the original Day-After-Tomorrow arrangement has unfortunately disappeared, sorry. But you can still say that it played an important role as the starting point for the development of the song.

I will also thin out the instrumentation at the ends of the individual sections to make the transitions clearer. Then I want to eliminate the second octave jump in the melody in the last section (@Wave Mechanic's suggestion). 

Additionally, I plan to add vocal effects to the currently purely instrumental section and maybe also at the final section. These would come from the singer this song is about. That's a bit tricky, because I don't want to use material from her songs directly (so as not to run into any copyright issues). So I only have a few smartphone recordings of poor quality that I think I can use for this purpose. Finally a reasonable use case for all this audio restoration stuff that's here on the computer.

When it's finished, I will post the result in the blogs section. But it will certainly take a few more days (hopefully not weeks).

Not sure about that. I think, this thread is already too long for that. New posts would end up somewhere down in Nirvana.

By the way, I made good progress with the song today. Although I've had the Zynaptiq Pitchmap effect for a while, this is actually the first time I really use it. It's great, you control the plugin with a harmonic progression, for example by an Autochords track from Synfire, and it automatically adjusts incoming audio material on the assigned track of your synchronized DAW harmonically. In this way, you can move audio snippets around just like figures in Synfire, they always fit in harmonically. Cool. But because it's so cool, I'm going to keep working on the song for a while.