How to make back vocals sound more powerful and deep?

Question

Link to an mp3 with an audio material

I'm interested in part that started from 03m:05s.

Ideas that I have:

1. Overlaying same recorded back vocal over each other multiple times.
2. Overlaying different vocal parts sang with different pitches (like one part (original) sang in original pitch, second stretched up by a major third, and pitch up the same part by perfect fifth and then mix them together).
3. Add 40ms delay to make vocals sound "multiple".

Any other steps that I may forgot or could you guys please share any tips and tricks that I should know?

Will be very grateful if you tell me.

Answer 1

Unison - Detune

Simultaneous voices singing approximately the same pitch, but slightly detuned from one another. Inspired by the sonic artifact produced when a choir sings a note in unison without being in perfect tune with each other, this characteristic can be replicated by duplicating a single audio recording and moving the pitch randomly up and down within a certain pitch range. The more voices and the wider the pitch range, the more dramatic the effect. Also, playing simultaneous, but separate recordings of the same singer will naturally produce this effect since he/she will not sing additional takes in the exact same way.

Stereo Spread

Since you have simultaneous voices from the unison effect, panning each voice a little to the right or to the left will increase the "width" of the background vocals, yielding a wide stereo image.

Harmonic Doubling

In the provided example, the voices are harmonizing, with the most pronounced interval being the perfect 5th. This interval is special because it is one of the lowest harmonics in the overtone series and, as such, is so consonant with the fundamental pitch that it can be perceived as one and the same; however, it adds a second set of formants and beating patterns that makes the sound pick up a stronger and more aggressive timbre. In addition, voices can be "doubled" an octave above or below. This will also work to make the sound "taller" as it occupies a wider band of frequencies within the human range of hearing.

Conclusion

The result is very powerful because it occupies a large band of frequencies (thick/tall), doubles on very highly consonant frequencies (5th, octave), and has wider spacial imaging (stereo).

Answer 2

To add to James' answer…

I would personally never use a delay to fake the number of voices, I would always just keep adding tracks til I had enough.

Secondly, I always pan all BVs away from centre [spread from 10% to 100% width], leaving a hole for the lead vox if there is one, or just a space in the centre if there isn't. So long as it's not too big a hole it can really work well.
I also reduce the volumes of the vox in the centre compared to the sides, to try generate a 'horseshoe' shape to the stereo field. Having the centre 'further back' than the sides always adds to the perceived depth of the image.

I also tend to leave pop BVs totally dry, using the very slight room ambience to do its job.
I never auto-tune BVs either; it kills the spread effect you're trying to create.

For 'pop' I use a heavy multi-band compressor to keep the high frequencies loud but under control.

I've posted a couple of examples to soundcloud, very different feel, but same method used for both.

&

[Let me know if you don't see those as embedded soundcloud files - they're always a bit hit & miss for me on SE]