...try GAG (golden Audio Gate) by kjaerhus.
very versatile gate!
try also playing with attenuation, you might reach more natural sounding results, by attenuating the reverb, than by fully cutting it out.
i use the transient mod plug that comes with cubase...i'm sure spl, stillwell, flux, or whoever would work just as well. just got done mixing/mastering a spoken word project with too much room ambience on everything and it was a lifesaver. just turn the release down as far as you can without it sounding bad.
Is there any way to cut out as much of the natural reverb on the vocal track as possible? The vocal was recorded in a poor setting that picked up too much natural reverb.
There is no satisfactory way. You can use a gate but it'll make the vocals sound choppy. I suppose if you'd really good and don't mind maddeningly tedious work you could try hand editing in a DAW, redrawing the signal at the ends of phrases, but that won't do anything for the muddy sound DURING the phrase. And the ends still probably won't sound right.
So no, there is no satisfactory way. Recut the vocal or live with it.
Try using a multi-band expander/gate like the Waves C4 or McDSP ML4000 to expand/gate the high frequencies only. Most of the room noise exists in those frequencies, and the only part of the voice that exists there are transients. So a fast attack and release should let them poke through. You almost don't hear it working, if set up properly, and it's more transparent than a broadband expander. We use this technique on all the spoken word recordings we do, by default.
Listen through the recording to find words that are masked by the reverb. Loop this section of audio and with a wide Q eq, cutting around -6db, sweep around the area that appears to have the room build up until you hear the performance become clearer (there are usually two distinct bands for each identifiable build up). Reduce the amount of db cut as much as possible without losing the clarity you have gained (I suggest -3 to -4db - the aim is to reduce the perceived roominess without cutting too much of the signal as, the more you have to increase it to compensate, the more you bring up the room and general noise floor)
Now with narrow band eq's, go through and reduce the reasonant frequencies that are most attention grabbing in these areas. I would suggest not more than -7db as you start to create noticeable, adverse phase shifting. You could use a linear phase eq but I generally use apQualizr due to the amount of bands often required to fix these problems.
When doing this I always pull back a little with my cuts so as not to lose too much of the signal and when cutting I close my eyes so I can pay closer attention. A/B constantly to make sure you haven't gone too far and check you Q values.
I only cut with eq when doing this procedure
Volume map your levels up so you need to compress less (compressing post fader)
When compressing use slow release times so you don't expose the room so much at the end of words.
When volume mapping out the tails, start cutting before the word has finished so the transition isn't so obvious.
Use additional, intentional reverb to mask the embedded stuff
Use tone or hiss to mask some of the verb
There's a bunch of other stuff you can do searching for prolonged frequencies that don't move rhythmically in sequence with the performance but explaining is not so easy.
You can also look at modal room calculation but I find using my ears gets a better result.