/ɛ/as in bed

1. Say the word 'bet' with your normal, higher tongue position and notice how it feels.
2. Now open your jaw much wider than feels natural and drop your tongue very low, saying 'bet' again in an exaggerated, almost silly way.
3. Repeat the exaggerated version three times, feeling the big difference from your habitual position.
4. Gradually reduce the exaggeration by about half, saying 'bet' again with a smaller — but still noticeable — jaw drop.
5. Reduce one more time until the jaw drop feels natural but still clearly lower than your first attempt.
6. Compare 'bet' and 'bait' back to back, checking that 'bet' keeps the lower jaw position.
7. Record yourself saying both words and listen for a clear height difference.

Success check: You should feel your jaw and tongue settle in a clearly lower position for 'bet' than for 'bait', and the two words should sound distinctly different, not identical.

Exaggerated ɛ (overshoot)Jaw dropped much further than needed, tongue pressed low and flat, mouth very open — an intentional overshoot.Looks almost the same as the picture.

Target ɛ (dialed back)Jaw and tongue return to the normal, slightly-open mid-low position used in natural speech.

Why this works. Overshooting the jaw-lowering and tongue-lowering gesture for ɛ beyond the normal target, then gradually dialing back to the correct degree, creates a clearer acoustic contrast (higher F1) against a habitual mid-front e substitute; research on acoustic exaggeration in phonetic training shows this strengthens categorical perception and production before the exaggeration is faded.

1. Play the recording of 'bait' and 'bet' back to back, several times.
2. Close your eyes and just listen for whether the vowel glides upward or stays flat.
3. Write down 'A' or 'B' as you hear each word in a shuffled sequence of the pair.
4. Check your answers against the answer key.
5. Repeat the same listening task with 'wait/wet', 'late/let', and 'mate/met'.
6. Note any pair where you consistently guess wrong, and replay it five extra times.

Success check: You can correctly identify at least 8 out of 10 shuffled words across the four minimal pairs without seeing the spelling.

bait (eɪ)Tongue starting mid-high and gliding slightly upward toward the end.Looks almost the same as the picture.

bet (ɛ)Tongue staying lower throughout with no glide.

Why this works. Repeated AX/ABX listening on pairs that differ only in vowel height and glide presence trains perceptual categorization along the one acoustic dimension that gets collapsed, strengthening the learner's internal category boundary before production practice begins.

1. Read silently: 'I said the red-headed man gets many letters every week,' noting the underlined ɛ words.
2. Say the sentence slowly, dropping your jaw clearly on 'said', 'red', 'gets', 'many', and 'letters'.
3. Speed up to a natural conversational pace while keeping the same jaw opening on those words.
4. Record yourself reading the sentence at natural speed.
5. Listen back and mark any ɛ word that sounds too close to the eɪ vowel in 'day'.
6. Reread the sentence, exaggerating the jaw drop only on the marked words.

Success check: At natural speed, every targeted word ('said', 'red', 'gets', 'many', 'letters') has a clearly open, flat vowel with no upward glide.

ɛ targetJaw held open and flat through each underlined vowel in the sentence.

Why this works. Embedding the target vowel in full carrier sentences rather than isolated words gives the learner practice maintaining the lower, glide-free articulation under natural coarticulatory and rhythmic pressure, which training research shows produces more pronounced, durable gains than isolated-word drills alone.

1. Say 'bait' first, feeling your tongue glide slightly upward.
2. Immediately say 'bet', dropping your jaw more and keeping the tongue flat with no glide.
3. Repeat the 'bait–bet' pair five times, exaggerating the jaw drop on 'bet'.
4. Record yourself saying all four pairs: bait/bet, wait/wet, late/let, mate/met.
5. Play back the recording and check that each ɛ word sounds clearly lower and flatter than its eɪ partner.
6. Redo any pair that sounds too similar, dropping your jaw even further on the ɛ word.

Success check: A listener (or you, on playback) can clearly tell which word in each pair is which, based on jaw height and the presence or absence of a glide.

Produce baitTongue rising slightly from mid to high during the vowel.

Produce betTongue dropping and staying low and flat through the vowel.

Why this works. Producing the two members of each minimal pair back-to-back forces the learner to actively control tongue height and glide presence as the sole distinguishing gesture, converting the perceptual contrast established in listening practice into a reliable, repeatable articulatory routine.

1. Say 'cat' and notice how wide open your jaw drops for that vowel.
2. Hold that same open jaw position without moving it.
3. From that position, raise only your tongue body slightly higher, without closing the jaw.
4. Add voice and say 'bed' using this adjusted tongue height with the same open jaw.
5. Alternate saying 'cat' then 'bed' five times, feeling the small tongue-height change while the jaw stays equally open.
6. Record both words and compare your 'bed' vowel to a native model's ɛ.

Success check: Your jaw stays about as open for 'bed' as it is for 'cat', but the vowel sound is clearly higher and more ɛ-like than æ.

Proxy: æ (cat)Jaw very open, tongue low and front, already a familiar target.

Target: ɛ (bed)Same jaw opening as æ, but tongue body raised slightly higher.

Why this works. English ɛ shares its jaw-lowering gesture with the vowel æ (as in 'cat'), which Spanish-background learners typically already produce with an adequately open jaw; by starting from the æ jaw position and raising the tongue body slightly while keeping the same open jaw, the learner recruits the same jaw-depressor muscle activity already mastered for æ as a proxy motor target for ɛ.

1. Look in a mirror and open your jaw as if starting to say 'bed', but freeze in that open position.
2. Check that your jaw is clearly more open than when you say 'bait'.
3. Hold that open position steady for two full seconds without moving your tongue.
4. Now add voice and say a flat, held vowel for two seconds, keeping the jaw position frozen.
5. Shorten the held vowel gradually over several repetitions until it's a normal-length syllable.
6. Say the full word 'bed' at normal speed, keeping the same jaw opening you practiced.
7. Record the final normal-speed version and compare it to a native model.

Success check: In the mirror, your jaw is visibly more open on 'bed' than on 'bait', and the vowel sound stays flat with no upward glide even at normal speed.

Step 1: openJaw dropped to a clearly open position, wider than for the eɪ vowel.

Step 2: hold flatTongue held steady at that lower height with no movement toward a higher position.

Why this works. Slowing the vowel articulation into discrete steps — mouth-open position, tongue height check, steady hold — lets the learner consciously verify jaw opening and the absence of an off-glide before reintegrating the sound at normal speed, addressing the core articulatory error of insufficient lowering.