Vocals

[Entorwellian]

Vocals are one of the more precarious elements that someone can perform in an IDM track. Quite a few artists have ruined their careers just from adding vocals in, even if they are semi-good, and never recovered (Cex, Tycho, some other ones in the EDM community). Hell, Squarepusher was lucky to get out unscathed from doing F-Train on Do You Know Squarepusher. Even as an instrument, over-reliance of vocal elements in an IDM track make it sound disingenuous to the electronic part of the composition even if they're chopped up or whatever, which is why I can't really get into Holly Herndon's stuff either. I'm not tone deaf and I've written lyrics for other people, but I never could integrate them into my own music deliberately. It's something I really want to do, but I have never found a way to integrate them in a way that didn't sound like they were divorced from the track or have to turn them into a part of the harmony to hide their presence. I've shopped around for singers for my material but not one of them that I met could add anything additional my songs in a way that thought was appropriate or synergistic, and that either the singer was competing with the track or the track was competing with the singer.

 

It's something that I play around with in my mind and occasionally sing with some of my tracks off-the-record, but it feels off to add pop-lyrics into something technically inclined and it also feels off being pedantically obtuse to the point of being pretentious for egotistical reasons. Is it possible?!? I'm more just curious with some of the other producers on here if you've wanted to do vocals on your tracks but couldn't for technical reasons, confidence reasons, or whatever.

[Braintree]

I've done some in the past with mixed results. That was before I discovered where my voice sits, so I stopped doing an impression of Depeche Mode with skittery beats. Here's a remix I did for a folk singer:

I basically took her material (banjo/voice/violin) and turned it into an IDM tune. I think female vocalists tend to work better for this genre, but there are some gems out there (Venetian Snares' "Dad").

[modey]

I've used my vocals in the past (my last Pselodux album had me attempting metal vocals, which may have been a bit of a mistake; the vocoder parts turned out well though). I'd love to get better at singing though, as all of the good vocalists I know are good producers already, so i don't know what I'd bring to the table in a collaboration. I really want to write a shameless 80s synthpop album, as that's where my vocal range naturally falls—that lower register new romantic/goth style

[dingformung]

Voice transports personality. I just can't stand most people's personalities on such an intimate level that singing is. Just keep it out of your music. And if you sing, do it right. I lost a lot of respect for Venetian Snares because of his singing. His voice is good but I find it obnoxious to rub your own humanness right in another human's face. Singing is like Joe Biden's smile.

[Entorwellian]

1 hour ago, Braintree said:

I've done some in the past with mixed results. That was before I discovered where my voice sits, so I stopped doing an impression of Depeche Mode with skittery beats. Here's a remix I did for a folk singer:

I basically took her material (banjo/voice/violin) and turned it into an IDM tune. I think female vocalists tend to work better for this genre, but there are some gems out there (Venetian Snares' "Dad").

That's not bad at all

[Entorwellian]

50 minutes ago, darreichungsform said:

Voice transports personality. I just can't stand most people's personalities on such an intimate level that singing is. Just keep it out of your music. And if you sing, do it right. I lost a lot of respect for Venetian Snares because of his singing. His voice is good but I find it obnoxious to rub your own humanness right in another human's face. Singing is like Joe Biden's smile.

The Biden visual makes things worse 

I think you've got a point with voices bringing things to an intimate level. I think that is why vocoding and autotuning work as well as they do: they have a degree of separation away from the person singing it so they sound machine-like yet still being able to have a nuanced human expression. I wouldn't want to go down that route though because it complements the track, it sounds cheesy and artificial to the point that it can be identified by it's era.

[Zephyr_Nova]

The key is to make the personality of the voice match the personality of the sounds.  Tim Exile, Bjork, Broadcast, Portishead, and Radiohead are all artists who I think have blended interesting electronic music with vocals that benefit the music.  But they've all made electronic music with the vocals/vocalist in mind.  Your chances of finding someone who's going to know what to bring to your IDM tracks are rare... though it can happen.  But it is best to know the voice you're going to use in advance so that you can tailor the sounds accordingly.

[Zephyr_Nova]

I tried this on a noise release a few months back.  I wouldn't say the voice I used is representative of my personality, rather I was taking on the persona of an entity that would inhabit this particular musical realm:

https://themunicipality.bandcamp.com/releases

[thawkins]

The most vocals I have ever done is put a poem through a text to voice robot and set it to an ambient track. I think it worked out OK. There's other examples on that album too.

https://trenthawkins.bandcamp.com/track/intro-of-loving-grace

In my experience vocals work well in electronic music if they are do not feel grafted on in production. This means both that the mix should not suck and the vocals sound/personality/the story they tell needs to form a coherent whole with the rest of the song.

But I also feel that if you have worked on your music without any vocals for most of the time and then you try to bring some in - it will sound weird and out of place because you are not used to it. I am pretty sure it takes huge amounts of time, experimentation and fiddling around to get to a point where you both happy with the result and it sounds good to other people too.

I am of course just pulling this out of my ass but I have felt the same thing when "searching for the sound" in bands - it takes time and getting used to each other.

 

PS. that Biden comparison was uncalled for. This sort of thing ought to be against the rules. I had a great day so far...

[Squee]

I really want do more vocal stuff and have been in contact with a couple of vocalists but it hasn’t been that fruitful so far. I’m pretty happy with this demo though. 
 

 

[modey]

8 hours ago, thawkins said:

But I also feel that if you have worked on your music without any vocals for most of the time and then you try to bring some in - it will sound weird and out of place because you are not used to it. I am pretty sure it takes huge amounts of time, experimentation and fiddling around to get to a point where you both happy with the result and it sounds good to other people too.

Yeah, I think this is what I came to realise after I made that metal album. My instrumental production skills are way higher than my singing ability / vocal production skills. Sadly though, the only way to get past that mismatch is to make music with mediocre vocals for a while until I get better at it.

[YEK]

I agree with thawkins about using electronic voices. you could use a vocoder talk box text to speech, all that work well with electronic music. it doesn't sound human like a real voice but most music i listen to isn't very human.

[thawkins]

11 hours ago, yekker said:

I agree with thawkins about using electronic voices. you could use a vocoder talk box text to speech, all that work well with electronic music. it doesn't sound human like a real voice but most music i listen to isn't very human.

If you want to take it to the next level, you can train your own neural network and feed in text to it.

https://cloud.google.com/text-to-speech/docs/wavenet

[YEK]

1 hour ago, thawkins said:

If you want to take it to the next level, you can train your own neural network and feed in text to it.

https://cloud.google.com/text-to-speech/docs/wavenet

Very cool. Thanks thawkins.

[TheBro]

My vocals are in this one lol. 

 

[YEK]

 

 

[Braintree]

Feed some fucking beats into it for christ's sake.

[Goiter Sanchez]

I've been using my voice on my Sweguno tracks for a while now, initially because I didn't have a synth or samples to create a proper melody. I decided to hone the aesthetic and explore how my voice could work with the open-ended production & arrangement techniques that the IDMs allow for. I have a solid falsetto but actually just started vocal lessons so I'm excited to make my voice more consistent and expand my range. I also want to start messing with my voice more. There are many creative possibilities...

Here are some examples of my work:

 

On 2/29/2020 at 9:57 AM, Squee said:

I really want do more vocal stuff and have been in contact with a couple of vocalists but it hasn’t been that fruitful so far. I’m pretty happy with this demo though. 
 

 

Squee I ended up recording a bunch of vocals for an instrumental you sent me a long while back. I haven't opened it in so long I have forgotten exactly what I did but will have a listen and see if it's worth sending over!