Certain aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment may determine a predicted tune-away for a set of voice frames. In some aspects, the user equipment may alter a vocoder rate from a first vocoder rate to a second vocoder rate that is less than the first vocoder rate based at least in part on determining the predicted tune-away for the set of voice frames. In some aspects, the user equipment may drop, in connection with an occurrence of the predicted tune-away, a subset of packets for a time period corresponding to the predicted tune-away. Numerous other aspects are provided.