MIT's 9.35 Perception course continues with Josh McDermott covering how the auditory system parses complex sound scenes before turning to speech perception specifically. McDermott walks through how vowels and consonants are produced, how formant frequencies determine which vowel a listener hears, and how small acoustic changes shift perceived speech sounds. The lecture uses the Yanny versus Laurel illusion as a case study, showing how the same audio signal can be heard as two different words depending on which frequencies a listener's ear or playback equipment emphasizes. The session builds on prior auditory scene analysis material, connecting general principles of how we separate sound sources to the specific problem of decoding speech. Delivered as a standard lecture hall session with slides and audio demonstrations, it gives a grounded look at the acoustic cues underlying everyday speech comprehension.