Question I can't figure out a workflow to combine BGM with vocal synth singing? (Original song making)

[uncreepy]

I have been trying to accomplish my goal of making my first short original song using a vocal synth + a DAW (either Vocaloid, CeVIO, Synth V, whatever-- I just want tips from producers), but am intimidated and having a hard time finding a good work flow to combine the singing with the background music.

I have tried making the singing in isolation in the standalone version of vocal synths like Synth V and Vocaloid, where the singing is broken into parts for my ideas on the melody/lyrics, but obviously the spacing is wrong because there is no background music to time it to.
I have tried working on just the background music in Studio One (I also have FL Studio but am not very good at using it), but I don't get very far because I worry about "What if the timing or these notes has to change when I bring in the singing, what if I have to rearrange a lot of stuff for both the BGM and singing?".

I get some progress at each task by themselves, but am not sure how to combine them. Or WHEN to combine them? Am I supposed to write the BGM completely, and then put in the singing? Or do I work on the BGM a bit, and start adding singing so I can tweak both?

When Vocaloid or Synth V is linked to a DAW, you can only hear the singing if you play the audio in the DAW itself, you can't hear the singing in the vocal synth software like you can in their standalone versions (the playback bar just won't move until you press it in the DAW). It seems really slow and awkward having to test out how the singing is going between two programs, basically, and I quickly want to give up because it's not very easy to work with.

I have seen that the human voice sounds like a clarinet, so do producers plan the singing while making the BGM, but just use a clarinet or something like that, export those notes as a MIDI to import into Vocaloid (for example), and just replace that MIDI sound with words when they're finally ready for lyrics in the end?

tl;dr Please give me advice on when to switch between lyrics (vocal synth software) and BGM (DAW) and how to do it so it's not overwhelming. Thank you for your time.

 

[mobius017]

You and me both! (We're both trying to make our first originals, so we're facing a similar issue. Actually, I'd be really interested to see what others with more experience have to say.) So what I'll say is based on what I've read and some trial and error on my part so far.

I've put together guidance piecemeal from a number of places, and generally the workflow I've come to think is recommended is to put together a melody for the lyrics first, then add lyrics to that, and then create a background to support that. It's advised to do the melody for the lyrics before the lyrics because it seems to be believed that it's harder to write music to match human speech than it is to make human speech match music. Which makes some sense, since you probably talk much more frequently than you write music, so you should be able to come up with words that fit the rhythmic/pitch structure more easily. It's theoretically possible to do the lyrics first, but supposedly it's more difficult.
That said, being the difficult person I am, I'm a writer, so I can't seem to help but try to write my lyrics first. (From the stuff I've read, it's a relatively individual process anyway, so while there are things that musicians often find easiest, each one might do it somewhat differently. And if you weren't originally a musician to begin with (*raises hand*), I think you might be more inclined to work differently than many people who were.) To me, getting the message of a song correct is really important. So I think it'll end up being a back-and-forth process: I'll do lyrics, then write a melody, find spots where they don't fit and adjust one or the other, etc. Then the background comes, with maybe the same kind of back-and-forth thing. I think it'll be a lot like sculpting clay that way (not that I've done that very much, but that's the image that comes to mind)--push this way, adapt, correct, kinda gently smooth things out, etc.

Personally, I like to use MIDI for planning things out. So I make the melody for the lyrics, the melody for the instrumental, whatever, in Anvil Studio and then send that to the DAW/Piapro, etc. Not that it doesn't change after the fact (which vexes me, since I'd like to say I have a perfect baseline to go back to if needed), but it's at least a place to start from, and I can build separate pieces of the song to arrange that way.

At least, that's what I'm doing so far. Hope that helps. Like I said, I'd love to see how more experienced producers work, too.

 

[uncreepy]

My sister has been reading some lyric books, and some people write the entire song before even having music to put it to (I guess they are lyricists and not necessarily the person who writes the background music).

I personally don't care that much about the lyrics, because I want to make song versions in both English and Japanese, so I'm not married to the words. I've done some song collabs for Vocaloid where people let me write the Japanese lyrics (and also secretly covered some English Vocaloid songs in Japanese), but the MIDI notes were already there for me to fill in, and I had to rewrite the Japanese lyrics so many times to fit that I realized being married to specific wordings is a luxury.
A lot of song tests I've done are just la's that become words later on, so I mainly just try to figure out the melody.

I guess my number one problem is just physically combining the vocal synth software in the DAW or combining the vocal synth software with the exported .wav (whatever people end up recommending). And dealing with how the software works together (like how you can't play back the singing unless it's in the DAW).

 

[mobius017]

And dealing with how the software works together (like how you can't play back the singing unless it's in the DAW).

Yeah, that's weird. I'm sure Piapro lets me play back the audio with the play button, though I think I might've heard of that issue with Vocaloid 5. Just as a dumb/simple suggestion, if you click "Configure Audio Device" in Studio One, on the Audio Setup > Audio Device tab, is the "Release audio device in background" checkbox filled in? If that box isn't checked, the DAW basically takes over the computer's audio, so stuff like YouTube won't have any sound. (Though, for me, I still have issues with Anvil not being able to make sound with the DAW open.) That's the only thing I can think of, aside from Vocaloid 5/Synth V-specific issues, that would keep your plugins from being able to play back on their own.

 

[Kona]

Here’s my advice doing it.

Don’t worry about the melody. Write your lyrics in a text document first.
When you are making your song is when melody should come in.

Pick a key and start making your instrumental. Focus on timing here a lot so you can be certain, have your metronome on, have drums and notes fall on bars or just on the beat in general, and just make sure your key is consistent.

Once that is secure, load in some random instrument that you can hear between the song like a piano or a synth, and create notes as a vocal melody. Title this track Vocal or something for now.
Tip for create nice vocal melodies fast: have them fall with the chords, at least the first note should be a part of or extension of that chords. For example, C Manor chord CEG, start your melody on one of those, also time it woth primarily eighth notes as your smallest note as to not get into too complicated timing.

After you are done woth that, if you’re in Studio One, go to File> Save As and in the drop-fown, change it from Studio One song to MIDI.
Import the MIDI into your synth of choice, any synth should ask you which tracks to import, import your Vocal track. From there, start inputting the lyrics.
If you feel more comfortable, export the song too and just import it and ise it to help you. Just drlete it before exporting from your synth.

After that, go back to your song project and go to Song>Import File. Studio One imports from the song position you’re at, so make sure it is back at the beginning or wherever you want to start.

Another more...abstract way you could do it is using the VST version of a synth. You just see the measure nimber where you want to begin in your instrumental, then switch to the synths and input notes. That’s how I have done it for both my released songs. I input notes in Piapro Studio, following bars for timing, then I go listen to it in context. This could be more effective in a way, since you hear it with your singer and can make changes to anything at that moment.
If you want to hear the vocal soloed with this method, just make the track solo, but everything is better in context with all instruments on.

Really bottom line is getting your pieces together is just do not underestimate the tools you have with timing and key. The beats are not numbered to be ignored, use them to your advantage.

 

[mobius017]

Thanks @Kona! I think those suggestions might be helpful for me, too!

Pick a key and start making your instrumental. Focus on timing here a lot so you can be certain, have your metronome on, have drums and notes fall on bars or just on the beat in general, and just make sure your key is consistent.

Once that is secure, load in some random instrument that you can hear between the song like a piano or a synth, and create notes as a vocal melody. Title this track Vocal or something for now.
Tip for create nice vocal melodies fast: have them fall with the chords, at least the first note should be a part of or extension of that chords. For example, C Manor chord CEG, start your melody on one of those, also time it woth primarily eighth notes as your smallest note as to not get into too complicated timing.

I've actually wanted to share this for awhile--I've been working at studying music theory, and I found a video that I thought was really good about using a scale (and, by extension, the associated chords) to write a melody. It's like half an hour long, but I think it's worth the watch because it explains why scales are important for melody and how different notes in a scale can be better for making melodies that work well. And it's done by a University of Wisconsin professor, so it's basically like getting a half-hour class for free. :)