There are a few ways you could go about this. One way I've been experimenting with has been to use a pitch detector on the violin's input and apply this pitch to the waveform of another instrument. I have tried this with a trumpet's waveform with ok results, it needs tweaking though as at the moment the pitch detection isn't great and it picks up all the noise of the fingers on the strings and analyses its frequency. One way to get round this would be to apply a smoothing filter first so the short, irrelevant bits of sound aren't counted as important.

If that all sounds a bit complex, then you could continue down the filtering route. A horn's sound is 'smooth' because of its frequency content. I'm not sure how much you know about this sort of thing, but all sounds are made up of a number of sine waves at different frequencies. For a pitched sound, there are a set of harmonics based on the frequency of the sound you hear and its multiples. For example if the frequency you hear (called the fundamental frequency or first harmonic) is 440Hz, the sound will also contain frequencies at multiples of this, eg 880Hz and so on. There are usually other frequencies in the sound that are not integer multiples of the first harmonic (the pitch you hear). These add a 'noisy' quality to the sound, which can make it sound richer. Horns, however, have the most sinusoidal sounds of all the instruments, that is to say that most of the frequencies that make up the sound are integer multiples of the fundamental frequency. SO, after all that rambling, another way to get a smoother violin sound would be to boost the harmonics (frequencies at integer multiples) of the fundamental frequency.

I hope I have neither lost you nor patronised you with my explanation. If I lost you I'll have another go at explaining - but harmonics and frequencies are a topic widely covered on the web too if you want to search and find out more.