Examqo › Examqo library › Speech-language pathology and audiology › Hearing science
Consonant acoustics and speech perception: VOT, formant transitions, fricatives and hearing loss
Consonants are identified by three cues: voicing (mainly VOT), manner (silence, noise, nasal murmur) and place (formant transitions and the noise spectrum). Consonants are weak and high-pitched and carry most of the intelligibility, so they are the first to vanish in a high-frequency loss.
Why it matters in the exam
VOT is the time from release to the start of voicing. In English /b/ is short lag (about 0–25 ms) and /p/ long lag with aspiration (above about 40); in Hebrew /b/ starts voicing before the release (negative VOT) and /p/ is short lag. A worked example: release at 120 and voicing at 185 — a VOT of +65, long lag. VOT lengthens as place moves back (bilabial, alveolar, velar). In categorical perception, discrimination is best for two stimuli on either side of the boundary. After every closure F1 rises; F2 marks place, and F2 and F3 converging (the "velar pinch") marks a velar. A glide such as /j/ starts with a high F2 and a slow transition; approximants differ in F2 and F3, and American /ɹ/ is recognised by a very low F3 that Hebrew /ʁ/ lacks.
Fricatives come from turbulent flow at a constriction — aperiodic noise. /s/ has its energy mainly at 4000–8000 Hz and /ʃ/ at 2000–4000 Hz, because a longer cavity lies in front of it; so /s/ before /u/ is lower (anticipatory rounding — coarticulation), and an /s/ with a low spectral centroid sounds like /ʃ/. A voice bar at the bottom plus high noise is /z/. The nasal murmur is a strong low-frequency resonance with weak energy above it. About 100 ms of silence between /s/ and the vowel is heard as a closure (say → stay), and a long vowel before a final consonant signals that it is voiced. Vowels carry loudness, consonants intelligibility. In a child with a steep loss from 2000 Hz, /s/ and /ʃ/ are confused; on the Ling six-sound test, responding to all but /s/ points to missing high-frequency audibility. The SII (from 0 to 1) expresses the share of speech information that is audible. The McGurk effect: /ba/ heard and /ga/ seen — most people report /da/. In sensorineural loss understanding is poor even when speech is audible, because resolution is poor; and talkers raise their voices in noise (the Lombard effect).
Key points
- VOT: English short versus long lag; Hebrew voicing lead versus short lag.
- F1 rises after every closure; F2 and F3 mark place.
- /s/ 4000–8000 Hz, /ʃ/ 2000–4000 Hz.
- Silence — a stop; a long vowel — a voiced final consonant.
- Ling, SII and McGurk; in sensorineural loss — poor resolution.
Summary table
| Cue | What it marks | Value or example |
|---|---|---|
| VOT | Voicing in stops | English: /b/ about 0–25, /p/ above about 40; Hebrew: /b/ negative |
| F1 transition | Manner — stop | F1 rises after every release |
| F2 and F3 transitions | Place | F2–F3 convergence — velar; /j/ — high F2 and slow transition |
| Frication spectrum | Place in fricatives | /s/ 4000–8000 Hz; /ʃ/ 2000–4000 Hz |
| Silence and duration | Closure; voicing of a final consonant | 100 ms of silence — a stop is heard; a long vowel — a voiced consonant |
| Nasal murmur | Manner — nasal | Low resonance, weak energy above it |
| Hearing loss | Missing audibility and poor resolution | Ling: only /s/ missed — too little high gain; SII between 0 and 1 |
In the full version for subscribers
- A simple algorithm
- Memory aid
- Exam traps
- 5 review cards
- What matters for the exam
The full version is open to subscribers of: Speech and hearing (communication disorders).
Subscriber? Read the full topic Plans and prices 25 free questions
Sources
- Kent & Read, The Acoustic Analysis of Speech, 2nd edition (2002) — Consonants
- Lisker & Abramson, Word, 1964 — A cross-language study of voicing in initial stops
- ANSI/ASA S3.5, 1997 (R2017) — Methods for Calculation of the Speech Intelligibility Index
Sources are given for further reading; the text is written afresh and not copied from them.
Reviewed by the Examqo AI Professor · 2026-11-02. The text may contain errors; when in doubt, check the source.