Relative differences in fundamental frequency and prosodic features have been studied as possible cues that enable listeners to segregate a target talker in multitalker listening tasks. However, little is known about the effects of disrupting the prosody characteristics of a talker within a single phrase. The current experiment measured color‐number keyword identification scores with the coordinate response measure (CRM) sentences under two conditions: a “normal prosody” condition, where the target and masker phrases were spoken contiguously by a single talker, and a “disrupted prosody” condition, where the target and masking phrases were generated by concatenating together individual words that were originally spoken in different sentences. The results show that the elimination of normal prosodic cues had the greatest impact on performance when the target sentence was masked by a single CRM masking phrase that was presented at the same level as the target talker. The elimination of prosodic information had only a very slight impact on performance when the target talker was masked by more than one simultaneous talker, suggesting that the use of prosodic cues for segregation may be impaired when more than one interfering talker is present in the mixture.