'Voice' is a musical term that has a couple of different meanings. In general terms a voice is one sound out of many. So for example in an orchestra a violin could be a voice, as could a flute, as could a trumpet and so on. In notation, a voice is one set of notes when different layers of parts are used. For example the part below has one voice:
Where as this part has two voices:
In the second example, all notes in 'voice 1' have there stems pointing upwards and all notes in 'voice 2' have them pointing down. This is a technique that is extremely common in drum kit notation but different people utilize it in different ways.
One common approach is to write the hands in one voice and the feet in a second voice. In a groove that would give you something like this:
The advantages of this method are that it clearly shows what each set of limbs is doing but the big downside is that it can very quickly become cluttered when more complex kick patterns are used. For example, look at this extract:
The additional rests that need to be used makes the part look far more complicated than it really is.
When notating fills this approach is also a bit hit and miss. Something like the extract below is great, you can see very clearly what you are playing with each limb:
However, when it comes to a fill like the one shown below the notation makes things look overly complicated:
The second approach is to voice the parts based on what job each voice is performing, this is usually a 'time keeping' part and a 'rhythmic' or 'melodic' part. Let's take another groove as an example:
In this pattern the hi hat (or riding) part is written in voice 1 with the kicks and snares in voice 2. So the part that is 'keeping' time is written separately to the parts that add interesting features to the groove. The advantage to this form of notation is that it shows more clearly what rhythms are being used in a pattern without needing too many rests. The downside is that the relationship between hands can get a bit lost. In the example below the notation on the left shows more clearly what the hands are doing than the notation on the right:
When it comes to fills the notation for this approach is often the same as the first. It could be argued that the first fill example discussed above that had a roll around the kit and some feet under it actually uses the same logic as this second approach would. The feet are performing a different job to the hands, so they are written in different voices. For the second fill example that had snares, crashes and kicks this approach to notation would group the snares and kicks in one voice and the crashes in another, giving you this:
To me this way of notating this part is far clearer and cleaner.
When notating it is entirely down to you which approach you take, you should go with the one that makes most logical sense to you for the part you are transcribing. Both methods are very common and both have there pro's and cons. I personally prefer the second method as to me it is a more 'musical' approach that is based more on rhythm and sound than the mechanics of which limb is doing which. I do however use the first method, particularly when notating 16 Beat Grooves or rolls around the kit that have some form of constant foot pattern underneath.
