In phonetics, aspiration is the strong burst of air that accompanies the release of some stop consonants. To hear and feel the difference between the aspirated and the unaspirated sound, put your hand in front of your mouth and say top and then stop. The t in top is aspirated; in stop, it is unaspirated.
Englishvoiceless stops are aspirated when they begin a stressed syllable (as in pen, ten, Ken), but this aspiration is not distinctive (They also have unaspirated variants in other positions). In many languages, such as Cantonese, Hindi/Urdu, Icelandic, Korean, Mandarin, and Ancient Greek, /t/ and /tʰ/ are different phonemes altogether.
Alemannic German dialects have unaspirated fortis /p t k/ as well as aspirated fortis /pʰ tʰ kʰ/; the latter series is usually viewed as clusters. Icelandic has pre-aspirated /hp ht hk/; some scholars interpret these as clusters. /b d g/ are normally also voiceless in Danish and most Southern varieties of German. Traditionally, they are still transcribed as <b d g>, even though what distinguishes them from their "fortis" counterparts <p t k> is mainly their lack of aspiration.
In the Unaspirated T, the voicing kicks in the moment the T is released, with no delay (or “Voice Onset Time”).
In Korean, for instance, they have different letters for each T/D sound (aspirated, unaspirated, and voiced), as well as for K/G and P/B. Thus, the difference between an unaspirated T and a D could be the difference between two completely different words (like “Cake” and “Bake” in English).
It seemed as though she did in fact contrast what I’m pretty sure is an unaspirated T with D (”say TaTa” versus “say DaDa”), and it seems like she really does see the sounds as different.
The k in key is aspirated; in ski, it is unaspirated.
English voiceless stops are aspirated when they begin a stressed syllable (as in pen, ten, Ken), but this aspiration is not distinctive (They also have unaspirated variants in other positions).
Alemannic German dialects have unaspirated fortis /p t k/ as well as aspirated fortis /pʰ tʰ kʰ/; the latter series is usually viewed as clusters.