Writing For Another Voice
October 12, 2010 11:02 AM
Does anyone have any tips for writing songs that work well for another person's voice? I've always just kind put songs together as they come and figured we would just adjust the key if it was difficult for our singer, but I think now that is kind of short sighted. Our singer has a good voice, and our music would be better if the songs really showed that off.
I'm interested in technical issues like how to determine the best keys for someone to sing in, but I'd also like to hear stories about how you have approached writing songs for others to sing and the challenges you faced. And if you'd like to make suggestions based on work we have already done, there are a lot of songs by my band posted here.
posted by InfidelZombie (10 comments total) 1 user marked this as a favorite
posted by InfidelZombie (10 comments total) 1 user marked this as a favorite
I've never really written for another person's voice, but I have tried to get better at being tactical in writing for my own voice, and part of that has been learning to think about what I can and can't do well and being a bit studious about my vocal range and the various uneven curves of expressiveness and power across that range.
So, some of the things I think about:
- Practical range. I can put out a steady, strong note up to about G4, the G above middle C; my range bottoms out around G2. A couple octaves is a lot to work with, but! Lots of caveats there because the strength of my vocal output varies a lot. If I want loud strong stuff, the octave between G3 and G4 is really all I've got, anything below that starts to get steadily more breathy and weak even when I push it. The last half octave or so down there really doesn't have any punch to it at all, I can sing the note but I can't sell you anything with it if it's rock music.
- Wear and tear. I can do G4 pretty confidently, but it's going to tire my voice out quicker than stuff a few notes down. I'm not gigging, so losing my voice a little doesn't have much practical effect on my life most days, but if you're playing out that's one thing to consider. A song that sounds great but pushes the singer's voice to the hilt in a "can't do this every night" sort of way may be a problem.
- Sweet spots and weird bits. Aside from just generally writing for the strong regions of my vocal range, I try to match up where my voice really sounds good with the bits of a melody that should be like Oh Hell Yeah moments. If sustained notes sound particularly solid when I'm singing e.g. an E4, I'll consider orienting the key or some bit of the melody around that note if possible. And while my comfortable range stops at about G4, I can reach up to an A4 or even a Bb4 sometimes if the situation is right—so I know that if I'm careful about it I can sneak that little trick into a song even though it's not a sustainable note. I can also manage a little bit of a controlled crack to my voice, though I don't do that much; but if there's a song where that little breakup of the vocal melody would sound good, it's nice to know I can go there.
- Room for harmony vocals. If I'm writing a melody with the expectation of harmony vox going with it, I need to think about where those harmonies are gonna go and if those will be singable or will otherwise work well with the melody I'm writing. I'm basically always writing harmony lines for myself, so the obvious implication there is that if I want a harmony line that sits on top of the melody, I have to make sure the melody doesn't stray to the upper limits of my vocal range. That means keeping a melody down to a like D4 or E4 ceiling, usually. If I'm gonna harmonize below, I can go high with the melody but need to think about not taking it too low or the harmony vox will end up in that breathy insubstantial part of my range and not really mesh convincingly with the melody.
- Harmonic context in the song. Vocals in isolation are one thing, vocals against a backing track are another entirely. If I'm recording a quiet acoustic song, I can pretty much use my whole range since my vox will still be fairly present; if I'm recording a rock song, I have to think a lot harder about how things are gonna work. If I want to use my low range in a rock tune, that means deciding to have the guitars drop a whole lot for those bits, for example, to evacuate that low-midrange area so my voice will be present and intelligible. If I need vox and guitars to both be going at the same time, I'm gonna write to the high end of my range so my voice has a chance of standing out in the mix.
- Picking a key the band can play in. This is one thing that I personally worry less about these days than I used to, and not having to worry about that is very freeing. It's meant me getting a lot more comfortable playing in awkward keys, of course—E Flat on the guitar, or F Sharp on the piano, stuff that doesn't make my stupid fingers happy—but the payoff is never having to feel like the song would be better if I could just play it in the key that fits my voice.
Those are most of the things I think about when I'm trying to work a melody strategically. Often I'll have the melodic bit I want in mind already, but I'll end up shifting it up or down by some number of steps to better make it suit the practicalities of my voice and the sort of arrangement I'm considering. In some cases I'll kill my darlings and rewrite part of the melody considerably to find some compromise between what I'd like it to be in abstract and what is actually gonna function in practice.
posted by cortex at 2:54 PM on October 12, 2010
So, some of the things I think about:
- Practical range. I can put out a steady, strong note up to about G4, the G above middle C; my range bottoms out around G2. A couple octaves is a lot to work with, but! Lots of caveats there because the strength of my vocal output varies a lot. If I want loud strong stuff, the octave between G3 and G4 is really all I've got, anything below that starts to get steadily more breathy and weak even when I push it. The last half octave or so down there really doesn't have any punch to it at all, I can sing the note but I can't sell you anything with it if it's rock music.
- Wear and tear. I can do G4 pretty confidently, but it's going to tire my voice out quicker than stuff a few notes down. I'm not gigging, so losing my voice a little doesn't have much practical effect on my life most days, but if you're playing out that's one thing to consider. A song that sounds great but pushes the singer's voice to the hilt in a "can't do this every night" sort of way may be a problem.
- Sweet spots and weird bits. Aside from just generally writing for the strong regions of my vocal range, I try to match up where my voice really sounds good with the bits of a melody that should be like Oh Hell Yeah moments. If sustained notes sound particularly solid when I'm singing e.g. an E4, I'll consider orienting the key or some bit of the melody around that note if possible. And while my comfortable range stops at about G4, I can reach up to an A4 or even a Bb4 sometimes if the situation is right—so I know that if I'm careful about it I can sneak that little trick into a song even though it's not a sustainable note. I can also manage a little bit of a controlled crack to my voice, though I don't do that much; but if there's a song where that little breakup of the vocal melody would sound good, it's nice to know I can go there.
- Room for harmony vocals. If I'm writing a melody with the expectation of harmony vox going with it, I need to think about where those harmonies are gonna go and if those will be singable or will otherwise work well with the melody I'm writing. I'm basically always writing harmony lines for myself, so the obvious implication there is that if I want a harmony line that sits on top of the melody, I have to make sure the melody doesn't stray to the upper limits of my vocal range. That means keeping a melody down to a like D4 or E4 ceiling, usually. If I'm gonna harmonize below, I can go high with the melody but need to think about not taking it too low or the harmony vox will end up in that breathy insubstantial part of my range and not really mesh convincingly with the melody.
- Harmonic context in the song. Vocals in isolation are one thing, vocals against a backing track are another entirely. If I'm recording a quiet acoustic song, I can pretty much use my whole range since my vox will still be fairly present; if I'm recording a rock song, I have to think a lot harder about how things are gonna work. If I want to use my low range in a rock tune, that means deciding to have the guitars drop a whole lot for those bits, for example, to evacuate that low-midrange area so my voice will be present and intelligible. If I need vox and guitars to both be going at the same time, I'm gonna write to the high end of my range so my voice has a chance of standing out in the mix.
- Picking a key the band can play in. This is one thing that I personally worry less about these days than I used to, and not having to worry about that is very freeing. It's meant me getting a lot more comfortable playing in awkward keys, of course—E Flat on the guitar, or F Sharp on the piano, stuff that doesn't make my stupid fingers happy—but the payoff is never having to feel like the song would be better if I could just play it in the key that fits my voice.
Those are most of the things I think about when I'm trying to work a melody strategically. Often I'll have the melodic bit I want in mind already, but I'll end up shifting it up or down by some number of steps to better make it suit the practicalities of my voice and the sort of arrangement I'm considering. In some cases I'll kill my darlings and rewrite part of the melody considerably to find some compromise between what I'd like it to be in abstract and what is actually gonna function in practice.
posted by cortex at 2:54 PM on October 12, 2010
So, to put that to you: Red's got a great voice, a lot of badass husky power that's part of why I like your guys' stuff so much. There's a definite spectrum of vocal power moving up and down her range; if you're trying to write for her more specifically than you have previously you probably want to start thinking more explicitly about where her strengths are for evoking any given sort of feel, and think about your melodies in terms of how they can exploit those strengths.
I'd say that Lay It On The Line carries her vocally about as far down as you guys can really go and have it work in the loud rock context you exist in; vox are there but it's scraping a bit and if the melodic return to the root of the chord she's doing there didn't work thematically with the lyrics I'd probably complain a little more. So that's one benchmark; F#3 or G3 is probably a sort of lower limit for most things and maybe avoid going down that far in general practice. It'd be interesting to see how that sounds up like two or three semitones by comparison. (Another experiment would be to really cut back on the guitars a lot for the vocally low bits, but that's so much of the song that it might take some doing to not feel weird.)
By comparison, Serious Song has her spending most of her time between B3 and E4 where there's a lot more kick, and the bits where she dips down to like A3 add some nice husky texture as a highlight on the low end instead of being a whole bunch of the melody. Flatters her voice better, I think. And she pops up to like B4 area a few times which feels fuckin' great as a bit of showy rock attitude.
Same Thing also works really well vocally with a low-on-the-verses, high-on-the-chorus dynamic; you get that sort of menacing growl from her down around A3 in the verses, then the blast up to shouty high belting and some nice little vocal squeaks and tweaks around the edges while she's working that.
posted by cortex at 3:09 PM on October 12, 2010
I'd say that Lay It On The Line carries her vocally about as far down as you guys can really go and have it work in the loud rock context you exist in; vox are there but it's scraping a bit and if the melodic return to the root of the chord she's doing there didn't work thematically with the lyrics I'd probably complain a little more. So that's one benchmark; F#3 or G3 is probably a sort of lower limit for most things and maybe avoid going down that far in general practice. It'd be interesting to see how that sounds up like two or three semitones by comparison. (Another experiment would be to really cut back on the guitars a lot for the vocally low bits, but that's so much of the song that it might take some doing to not feel weird.)
By comparison, Serious Song has her spending most of her time between B3 and E4 where there's a lot more kick, and the bits where she dips down to like A3 add some nice husky texture as a highlight on the low end instead of being a whole bunch of the melody. Flatters her voice better, I think. And she pops up to like B4 area a few times which feels fuckin' great as a bit of showy rock attitude.
Same Thing also works really well vocally with a low-on-the-verses, high-on-the-chorus dynamic; you get that sort of menacing growl from her down around A3 in the verses, then the blast up to shouty high belting and some nice little vocal squeaks and tweaks around the edges while she's working that.
posted by cortex at 3:09 PM on October 12, 2010
Thanks for the feedback guys, this is very helpful. It's interesting that you mention getting her involved in writing the material, Major-- the reason I posted this question is Red spoke to me last week about just that, since she has been going through some tough times she'd like to get out in song. To do that right I know we need to approach the material differently.
Her plan was to just give me her diary, but I think I talked her out of that. Better for her to pick the parts that mean the most to her, that have a poetic quality to them, and we can build the songs from there. We're going to schedule some songwriting time together, so I wanted to work up a few ideas we could use as a starting point.
Cortex, everything you've written here has given me a lot to think about in that regard. The stuff on your own process is really great. And I think I understand now why a song like Carry On has never really worked that well-- as cool as that open low E sounds on the guitar, it is just too low for her to sing that with any power in her voice. A lot to digest there.
posted by InfidelZombie at 1:11 AM on October 13, 2010
Her plan was to just give me her diary, but I think I talked her out of that. Better for her to pick the parts that mean the most to her, that have a poetic quality to them, and we can build the songs from there. We're going to schedule some songwriting time together, so I wanted to work up a few ideas we could use as a starting point.
Cortex, everything you've written here has given me a lot to think about in that regard. The stuff on your own process is really great. And I think I understand now why a song like Carry On has never really worked that well-- as cool as that open low E sounds on the guitar, it is just too low for her to sing that with any power in her voice. A lot to digest there.
posted by InfidelZombie at 1:11 AM on October 13, 2010
A pleasure to be of help IZ. If I was in your shoes, I'd approach this as if I was teaching someone to swim. Red may never have done anything like this before, she may be a little fearful and feel exposed. If she's gets thrown in the deep end she'll panic and maybe go under. You have to put her in an environment where she feels safe and supported - where she can, metaphorically, touch the bottom of the pool and has a couple of floats attached to her. I'd start by picking one of your songs that you know works with her voice tonally and melodically, and suggest to Red "Hey, let's try and write some new words to that tune you really sing well". Take it from there - little steps. Before you know it, she'll be jumping off the high board. Like everyone has the inherent ability to swim, I think everyone has the capacity to create - just has to be brought out in the right way. Good luck to both of you and I look forward to hearing the results eventually!
posted by MajorDundee at 4:42 AM on October 13, 2010 [1 favorite]
posted by MajorDundee at 4:42 AM on October 13, 2010 [1 favorite]
I like that approach, though I think I'll modify it a bit to tackle a new song we've been working on that she sings pretty well, but is really clunky lyrically. I'm a little paranoid about making too many changes to existing songs because the only bad experience I've really ever had on stage came while we were reworking Same Thing-- tried to play it one night three times, and each time someone different reveted back to the old arrangement partway through and the song fell apart. It totally destroyed the rest of our set-- it's the only time I've ever heard "you suck!" from an audience, and I hope it is the last.
If you listen to the original version I think the changes made it a better song, but I'm a little gunshy about going down that road again. Better to just write a new song that does what I wish the old one would have.
posted by InfidelZombie at 12:05 PM on October 13, 2010
If you listen to the original version I think the changes made it a better song, but I'm a little gunshy about going down that road again. Better to just write a new song that does what I wish the old one would have.
posted by InfidelZombie at 12:05 PM on October 13, 2010
Late to the thread, but wanted to say I wrote for years songs for somebody else to sing. While an adequate vocalist myself, I was certainly not a performer with pipes like my band's lead singer was. I found it quite liberating to be able to have musical ideas expressed by this other voice. The soul can be wailing, so that's the mindset you need to have, you need to imagine hearing what you are envisioning in the voice of the singer. I suppose you really need to know that voice first and the person from whom it emits, their best keys, the best tempo for their natural phrasing, etc. You want lyrics that the singer can personally relate to or at least understand in a visceral way.
Fun example of writing for other voices: Quincy Jones, Michael Jackson, Stevie Wonder and Lionel Ritchie teaching the rest of the singers "We Are The World".
posted by bonefish at 11:30 PM on October 13, 2010
Fun example of writing for other voices: Quincy Jones, Michael Jackson, Stevie Wonder and Lionel Ritchie teaching the rest of the singers "We Are The World".
posted by bonefish at 11:30 PM on October 13, 2010
I've written a lot for other people and I think the main thing apart from writing in a key which brings the melody into the singer's range is to write songs which the singer enjoys singing. By which I mean if the singer really loves to belt a chorus, write a belting chorus. If s/he loves smart wordplay, use smart wordplay. And if the song is personal, make sure it isn't so personal to you that it feels like they're singing about someone else. When you're singing, you really need to find something in the song to glom onto, whether it's a hook that you know you can tear the balls off, or a couplet you can snap out as if you just thought it up. It makes all the difference.
posted by unSane at 8:21 AM on October 20, 2010
posted by unSane at 8:21 AM on October 20, 2010
(And as Cortex implies, it's really useful to get competent at transposing into different keys. You can usually find a key in the right neighborhood that is not impossible to play. Eb on the guitar is never fun but E and D are not far off. Capos are very useful. Pretty much the first thing I do when I write a song is try it out in a bunch of keys, not just to find the sweet spot for my singing but also because sometimes the chord voicings are spectacular in a particular key and only so so in the original.)
posted by unSane at 8:24 AM on October 20, 2010
posted by unSane at 8:24 AM on October 20, 2010
Red loves to belt it out, that’s for sure (and she’s damn good at it.) She also loves lyrics with some emotional resonance to them; she is a very social person and that connection with people and their experiences is very strong for her. That’s not always easy for me to write, as I tend to be more disconnected from people. I’m more of an observer, but I don’t think that works lyrically for her; she needs to be part of the action.
Transposing is kind of a weird thing for me—if it is a punk song it’s no problem, I’m playing barre chords anyway so we can move it anywhere we want and I’ll work it out. But anything more complex is an issue, mainly because I don’t practice as much as I should so I don’t have the skills to transpose what I came up with. I have a capo but it doesn’t work well with my Tele, I should get one that fits a 7.25” fretboard radius.
posted by InfidelZombie at 11:56 AM on October 20, 2010
Transposing is kind of a weird thing for me—if it is a punk song it’s no problem, I’m playing barre chords anyway so we can move it anywhere we want and I’ll work it out. But anything more complex is an issue, mainly because I don’t practice as much as I should so I don’t have the skills to transpose what I came up with. I have a capo but it doesn’t work well with my Tele, I should get one that fits a 7.25” fretboard radius.
posted by InfidelZombie at 11:56 AM on October 20, 2010
You are not logged in, either login or create an account to post comments
But I'm aware that that kind of armchair philosophy isn't particularly helpful to you. I'll have a listen to your band and see if I can offer something more practical. One thing I'd suggest for starters - if you don't already do it, involve your singer in the writing of the material. Encourage him/her to contribute ideas - melodically or lyrically. Don't just write stuff and present it to her/him and expect it to work when they have no creative stake in it. That technique will also kind of naturally resolve the "which key" problem as well.
posted by MajorDundee at 1:46 PM on October 12, 2010