When the audio for "ba" is presented along with video of an actor pronouncing "ga", the consonant half-way between ("da") is perceived.
Japanese subjects had a tendency to produce sounds continuously whereas Thai subjects paused at the end of every sound; the pause between the first vowel and the consonant was longer than that between the consonant and the second vowel.
It may also be the dropping of a consonant between vowels, as in the word o'er for over.
The consonant /v/ has disappeared when between /u/ and /i/ or /a/.
Moreover, there appears to be a relationship between consonant realisation and vowel realisation, with vowels preserved more often when the consonant is lenited, and elided more often when the target for the consonant is fully achieved.
In Classical Sumerian, the contrast between the consonants b, d, g, z and p, t, k, s was not between voiced (with vibrating vocal cords) and voiceless consonants (without vibrating vocal cords) but between consonants that were indifferent as to voice and those that were aspirated (pronounced with an accompanying audible puff of breath).
In initial position, Eskimo uses only a single consonant; between vowels at most two.
A sequence of consonants was used to name the taxa at each tier alphabetically with "A" and "E" inserted between the consonants of tiers 2 and 3 and "OZEM" appended after the consonants of tier 1.
The main difference between the consonants in pea and bee, when these words are said in isolation, is not that the one is voiceless and the other voiced, but that the first is aspirated and the second is unaspirated.
The steps that correspond to the large articulatory movements between the consonants are heard as being much larger than the equal size acoustic steps that do not correspond to articulatory movements occurring in the listener's speech.
