Most research on timbre perception has studied isolated tones. This study compares timbre discrimination of isolated tones with discrimination in various musical contexts, both single‐voiced and multivoiced. Twelve different contexts were used (four isolated tonal comparisons, four single‐voice musical patterns, and four multivoice patterns). Listeners judged whether the timbre remained the same or changed during the trial. Two possible versions of any instrumental timbre differed in the physical information used in thier synthesis. Three instrumental timbres were tested in all contexts: clarinet, trumpet, and bassoon. The effects of context upon discrimination varied across instruments. The clarinet and trumpet versions were best discriminated in isolated contexts, with discrimination progressively worse in single‐voice and multivoice patterns. The bassoon versions were best discriminated in the single‐voice patterns, with equal discrimination in the isolated and multivoice cases. It is suggested that these results were due to pronounced physical differences observed between the spectra of the two versions of the bassoon that were not apparent between the versions of the clarinet or trumpet.