Homophones are words that share the same pronunciation regardless of spelling (e.g., missed-mist, peak-peek, rays-raise, tolled-told)
Homophones can be useful examples because they allow students to perceive pronunciation while overcoming interference from spelling. For instance, to illustrate that -ed can be pronounced as /t/, an instructor might present the homophone pair missed-mist to provide a point of comparison.
When teaching final -s as /z/, examples like days-daze, doze-does (i.e., female deer), grays-graze, gays-gaze, Oohs-ooze, pries-prize, and sighs-size can be useful. This comparative technique can be especially effective when incorrectly selecting /s/ produces an unintended word. For example, if a student incorrectly cues /llllllllluo͞oueuo͞oueuo͞ouesssssssss/ for the word Lou's, the instructor could simply provide the contrastive pair: lose and loose and ask, "Which words sound more similar?" and "What's different about this other word?" This helps them to discover (in spite of spelling) that Lou's is identical in pronunciation to lose, but different than loose. Other comparative trios include: Brews-bruise-Bruce, Grays-graze-grace. These triads offer opportunities for instructors to ask, "Which word is pronounced like grays? Graze or Grace?" While the student may have initially thought to cue the word with a final /s/, the contrast of graze and grace provides two clear anchor points to compare and, ultimately, realize that grays is likely to end with a /z/.
Students commonly request to their CS instructors, "Tell me what I say."
Case: A student in a class suggests that she uses a "soft z" and that her /z/ is much softer than the instructor's pronunciation and that it should therefore be cued as an /s/. The problem, again, is that we must not compare how one person sounds compared to another. The student must compare her /z/ to other sounds she makes. The instructor used a homophone pair to give her the answer. "Do you pronounce ice and eyes the same way?" The student replied that she pronounces the words differently. This tells the instructor that while the student may have a "soft /z/", it is still a /z/. Since she has a distinction between /s/ and /z/, she needs to cue that distinction no matter how soft or strong her version of each.
In parts of the Western U.S., like Utah, a vowel merger has occured and the distinction between /ɑ/ and /ɔ/ has been lost. The use of homophones can be useful in a class to suss out which students will use the chin placement for /ɔ/ and who will not. For example, by offering pairs like cot-caught, rot-wrought, knotty-naughty, stalker-stocker, the instructor can ask students, "Are these words pronounced the same or differently?" For students from the mid-Atlantic region, these words are likely to be pronounced differently. For cuers from Utah, Idaho, and parts of California, they might be pronounced the same.