- bʌt Primary stress
- fla͡ɪ Secondary stress
- ɚ Unstressed
- a͡ɪ Vowel
- k Consonant
How words are segmented
- Syllables
- Split by pronunciation, not spelling. fire is two syllables, /ˈfaɪ.ɚ/. Consonants between vowels go to the following syllable as far as English allows a syllable to begin with them (the maximal onset principle), so monster is /ˈmɑn.stɚ/. A stressed short vowel keeps the next consonant, because English syllables cannot end in one: rabbit is /ˈɹæb.ət/.
- Phonemes
- Each sound is one phoneme. Affricates /tʃ/ and /dʒ/ and diphthongs /a͡ɪ/, /a͡ʊ/, /ɔ͡ɪ/, /e͡ɪ/, /o͡ʊ/ count as one sound each.
- IPA
- Broad (phonemic) General American transcription. /ˈ/ marks primary stress and /ˌ/ secondary stress, placed before the syllable. In the syllable boxes, stress is shown by color instead: orange for primary, blue for secondary. Diphthongs carry a tie bar, as in /a͡ɪ/. Stressed and unstressed central vowels are distinguished: /ʌ/ and /ɝ/ when stressed, /ə/ and /ɚ/ when not.
- Nonsense words
- Made up from sound patterns English allows, with stress usually on the first syllable. Each one is checked against about 250,000 real spellings (including names) and every pronunciation in the CMU dictionary, so it is never a real word, whether written or said aloud. The spelling is only a suggested reading; the IPA is the target.
- Sources
- Pronunciations from the CMU Pronouncing Dictionary. Word list checked against the public-domain Moby Hyphenator to exclude names and abbreviations. Familiarity ranks from FrequencyWords (CC BY-SA 4.0). Dictionary pronunciations can differ from regional speech, so check a word before using it in an assessment.
About these pairs
- Minimal pairs
- Two real words whose pronunciations differ by exactly one sound, such as cat and bat. Both consonant and vowel contrasts are included. The contrast is matched by sound, not spelling.
- Maximal oppositions
- Minimal pairs whose two contrasting consonants are as different as English allows, differing on at least three of voicing, place, manner and sonorant class, such as /s/ against /m/ in sat and mat. Used in maximal opposition and empty set approaches.
- Vocabulary level
- Words are filtered by when children typically learn them, using age-of-acquisition ratings (Kuperman, Stadthagen-Gonzalez and Brysbaert, 2012), combined with a frequency cut so rare words like groat cannot slip through. Ratings are averages from adult raters, so treat a level as a guide, not a guarantee, and skim a list before using it.
- Position
- Initial means the contrast is the first sound of both words, final the last, and middle anything between. Position refers to sounds, so knee and bee contrast initially even though one is spelled with two letters.
About response elaboration training
- What it is
- A treatment for aphasia (Kearns, 1985) that shapes and lengthens whatever the person says, rather than teaching set responses. Nothing the person says is wrong; every response is built on.
- How to use these sheets
- Show a picture of the prompt, then work through the five steps. The wh-cues are suggestions; follow the person's own words. Model sentences are written to be expanded, not read aloud as targets.
- Pictures
- The original protocol uses picture stimuli. This page gives the prompts and cue sequence; pair each one with a photo or picture card of your choosing.
About sound production treatment
- What it is
- A treatment for acquired apraxia of speech that works through a fixed cueing hierarchy on a set of target words, moving to a stronger cue only after an error and back down as production improves.
- Check the steps against your source
- The five steps here summarise the hierarchy as it is commonly described (modelling and repetition, integral stimulation, articulatory placement cues, an orthographic cue, then repeated practice). Wording and criteria vary between published versions, so follow the protocol you are trained in.
- Word choice
- Pick a target sound and position, and the sheet draws real words containing that sound. Everyday words are the default, since treatment items are usually familiar and functional.
About SODA
- What it is
- A way of classifying speech sound errors into four types: substitution, omission, distortion and addition. The six patterns here are common phonological processes, each a kind of substitution or omission.
- How to score
- Present each word, write what you heard, then mark the error type. Totals at the foot of the sheet show which types dominate.
- Likely error
- A prediction of how the word sounds if the pattern is present, with the changed sounds highlighted. Cluster reduction keeps the least sonorous consonant, the most common outcome, but children vary. It is a guide to listen for, not a score.