I’ve had Synthesizer V for several weeks now. I’ve purchased four vocal databases – Natalie, Hayden, Astoria, and Saros – and they each have their pros and cons, much like the real life versions of those four singers. I already made a video about why Synth V (as us fans call it) was the best choice for me, but the more I use this software, the more I love it. Vocal takes at 2 in the morning? Sure. Want to change one note or one word in the middle of the day? No problem. Want to try the song in several different keys, instantly? Piece of cake.
Why do we say “piece of cake” for something easy? I’m far too lazy to Google it.
Today I’m working on a cover of one of my own songs, featuring Solaria. I thought Hayden might sing it, but ultimately I ended up choosing Solaria. And on one hand, you can get a rough demo going in almost no time. My current workflow is to take an entire verse and put in the “la la’s”, so to speak. Then I go in and add the lyrics. When the pitch and timing are correct, I start to fix the words that sound funny. Once everything sounds good, I put in the vocal inflections I want to hear.
When you draw in a note on the grid, Synth V automatically inserts “la.” Hence the la la’s. You can play the melody in with a MIDI keyboard, but honestly, I’m finding it easier to use the pencil tool. When I play it on the keyboard, I find I spend too much time fixing mistakes.
The other thing I do is to select groups of notes for verses and choruses in the la la stage, and copy/paste them. You might think that’d make them all sound the same, but it doesn’t, because you haven’t put in the words yet. And the words will determine some of the inflections.
I do have a love/hate relationship with what people call “AI” these days. On one hand, I refuse to use AI-generated art or AI-generated lyrics. On the other hand, I’m happy to let AI play the drums for me (which I will then tweak) or fine tune vocal synthesizer singers for me (such as with Synth V.)
The fact is, I’m 59 years old and losing my voice. And Synth V has given me four highly talented vocalists I can use. Some might say “They’re not real.” True – and the sampled drums I use aren’t real, and the digital reverb isn’t real, the sampled Rhodes isn’t real…
The Synth V vocal databases were created by real singers. If it helps you to think of them as sampled vocalists, think of them that way. All I know is that I was ready to quit songwriting, but Synth V gave it back to me. And now, instead of just my limited bass vocal range, I’m like Alan Parsons – I have my pick of talented singers. Yes, right now that group is “only” four different singers, but I’m sure there will be more eventually.
When we strip away the humanity of something and use a digital version, do we reduce the quality of the art? I think yes, we do. Synth V’s Natalie is great, but human Natalie is even better. The Jupiter 8 inside of the Arturia V Collection is great, but the actual hardware synth is better. And so on.
But even Alan Parsons doesn’t have an unlimited budget for making a new album.
And if I have the choice of either shipping a cool song with some of the humanity stripped away, or just quitting, I’ll take the compromised version of the music. It’s still good. It can still make you feel something. And I don’t like limitations.