Which two actions ensure voice quality in a branch location with a low-speed, high-latency WAN connection? (Choose two)

Prioritizing voice packets via QoS mechanisms such as Low Latency Queuing (LLQ) or DSCP marking ensures voice traffic receives preferential treatment in the output queue, reducing jitter and delay on the constrained WAN link.

Fragmenting large data packets using Link Fragmentation and Interleaving (LFI) prevents large frames from serializing ahead of voice packets, which on slow links can introduce serialization delay exceeding the 150ms voice quality threshold.

Replacing electrical links with optical links does not address the WAN speed or latency constraints, which are determined by the service provider circuit, not the physical medium at the branch.

Increasing memory on the branch switch has no effect on WAN serialization delay, queuing behavior, or voice quality across the WAN connection.

Increasing WAN bandwidth addresses capacity but is not a design technique for ensuring quality on an existing low-speed link - the question asks about actions taken within the constraints of that link.