Should convert 'k' and 't' sounds to 'g' and 'd' sounds when they follow 's' in a word for pronunciation?

For example, stamp is pronounced as 's dæm p', etc.

  • 5
    Did you check any dictionaries for pronunciation guides? Have you seen something that suggests that you should make this change?
    – stangdon
    Commented Jun 1, 2023 at 15:35
  • Could you please explain your question? I do not understand it at all.
    – Lambie
    Commented Jun 1, 2023 at 17:19
  • 6
    Hi. I have never heard a native English speaker pronounce stamp like that. Where did you get this information?
    – Billy Kerr
    Commented Jun 1, 2023 at 19:00
  • 2
    Perhaps the best thing would be to listen to the way that native English speakers pronounce words, and then pronounce them the same way. Commented Jun 2, 2023 at 10:26
  • 2
    @TannerSwett If you're on ELL, surely you know that phonemes sound different to people depending on their L1. Confusing unaspirated /t/ for /d/ is not that ridiculous. Commented Jun 2, 2023 at 17:45

6 Answers 6


No. You should pronounce it as an unaspirated [t].

Longer: There are two phenomena in play: aspiration and voice.

sounds voiced? aspirated?
/p t k/ no at the start of a syllable
/b d g/ yes no

Thus what you hear in a cluster like 'stamp' is the voiceless plosive /t/ without its aspiration. Its voicing does not change.

However, because many languages, probably incl. yours, do not use aspiration, English's use of it is very salient. Thus it's easy to mistake it for the quality that defines /p t k/.

But it isn't. It's secondary from the phonological point of view. The phoneme /t/ has phonetic realizations [t] and [tʰ] (among others); it does not have a phonetic realization [d]. Hence, it's [stæmp], but not [stʰæmp] and not [sdæmp].

Incidentally, English native speakers have a similar problem. Take away the aspiration and we think we're perceiving /b d g/. However, it's possible to train yourself to perceive the distinction.

Hence, it might be possible to argue that aspiration is the key element and voicing is the second. But even if aspiration is treated as the primary distinction, it doesn't really make sense to target a /d/. A /d/ should be impossible in a cluster after a voiceless /s/ from a phonotactics point of view.

  • 2
    Moreover the common romanisation of Chinese (Pinyin) uses "d" for "unaspirated unvoiced consonant", which causes all sorts of difficulty for English speakers learning Mandarin.
    – James K
    Commented Jun 1, 2023 at 17:21
  • 6
    Agreed. Short answer for Xuesong Gao: native English speakers can definitely hear the difference between 'stamp' and 'sdamp', and the second one is wrong. 'sd' is not an allowed consonant cluster in English.
    – stangdon
    Commented Jun 1, 2023 at 17:42
  • 3
    @stangdon Indeed, and one reason we can tell them apart is that there's no reasonable way (for us, anyway) to articulate a /d/ after an /s/, so we have to exaggerate the change in voicing and that makes it more salient than the aspiration change (again, for us). Commented Jun 1, 2023 at 18:58
  • 3
    @dan04 That's not really the same thing, though; in all of the words you list sd is across a syllable boundary, whereas in the hypothetical "sdamp" it isn't.
    – stangdon
    Commented Jun 2, 2023 at 14:53
  • 3
    @dan04 "eavesdrop" and "wisdom" have /zd/, not /sd/, despite the spelling
    – gotube
    Commented Jun 3, 2023 at 2:45

Yes, in fact. English speakers cannot tell the difference between "stamp" and "sdamp" (or "disgust" and "discussed", etc.) in most speech, no matter how much they tell you they can. It's possible to pronounce them so they can be distinguished, but this is not the same thing as doing so by default.

Linguist Geoff Lindsey demonstrates here that clipping the /s/ out of an audio clip of the word "speech" makes it indistinguishable from "beach".

  • 3
    Thanks for the downvote, care to comment? Commented Jun 1, 2023 at 20:40
  • 11
    I didn't downvote (and I wish people who did would comment), but I will note that our inability to distinguish between [t] and [d] wouldn't mean one should target /d/. The other language's realizations of /d/ may not actually be the same as our [t]. And as stangdon hinted, if we native speakers in particular try to clearly articulate a /d/ after an /s/, it's quite noticeably wrong. Commented Jun 2, 2023 at 1:34
  • 4
    @LukeSawczak And I'd argue that for non-native speakers, understanding that they're identical 99% of the time is more helpful than trying to hear a difference that doesn't exist. Understanding the fine nuances that occasionally make them different should come second. Thank you for engaging, genuinely. Commented Jun 2, 2023 at 2:21
  • 3
    I think the issue is with how one defines "native speaker", which Dr Lindsey conflates with "British or American". Native speakers of English in, say, India do not aspirate /p t k/ at all, in any position, and yet readily distinguish them from /b d g/. I spent my early life in India and realized pretty quickly after I moved to the US that I needed to aspirate if my American hearers were to hear my saying /pat/ as "pat" rather than "bat". Yet for me, that's code-switching. When I'm around other native speakers who grew up in India, I switch back to the unaspirated forms. (Contd.)
    – verbose
    Commented Jun 2, 2023 at 8:25
  • 3
    As for the examples in Dr Lindsey's video, in each set of examples, there was at least one person whose pronunciation made it abundantly clear to me that this was not (say) beach but peach, or door but store. Y'know what, Imma turn this into an answer.
    – verbose
    Commented Jun 2, 2023 at 8:29

It depends. To what accent do you, er, aspire?

Native English speakers from, say, India readily distinguish /p t k/ from /b d g/ in all positions of those consonants—even after f or s. That's due to the influence of Indic languages, where aspiration is phonemic. /p/ and /ph/ are meaningfully different: e.g., in Hindi, pal, पल, moment; phal, फल, fruit. Indeed, /ph/ is not phonemically distinct from /f/ in those languages. The following relations pertain:

pronunciation Indian listener American listener
[p] /p/ /b/
[ph] /f/ /p/

All this can make things a bit bewildering, as when the Indian speaker says "I gave the dog a pat" and the American listener wonders why the speaker gave the canine a flying rodent 🦇. Or when an American speaker says "You deserve some pats on the back" and the Indian listener ponders dorsal applications of butter, margarine, olive oil, etc.

To somebody from India, then, even after an s, /p/ is pronounced [p]. Or more precisely, it's pronounced the same way as /p/ is pronounced at the initial position of a morpheme. And that pronunciation is very distinct from [b] to an Indian ear, but not so much to an American one.

Dr Lindsey's video, referenced in @the-baby-is-you's answer, furnishes a case in point. For each pair of words (door/store, speech/beach) there is at least one speaker whose pronunciation does distinguish the pair clearly enough for my Indian ear. In general, though, he is obviously correct: US/UK pronunciations do not distinguish between /p/ and /b/ (etc.) following s or f. However, while Dr Lindsey conflates "native speaker" with "British or American", there are large numbers of native speakers of English elsewhere in the world. And I do mean native, those who grew up speaking the language either exclusively or in true bilingual/multilingual fashion, as opposed learning it as a second, third, or fourth language outside of an immersive context.

So if your native language isn't English, then does your native language give you the ability to hear and reproduce the distinction between [p], [ph], and [b]? If so, and if you don't mind having an accent that's somewhat distinctive, by all means reproduce those distinctions. If you cannot consistently hear or reproduce those distinctions, then by all means follow the advice to say "sbeach" or "sgool" for "speech" or "school".

At the end of the day, though, the point is to recognize that even native accents are bewilderingly various. Arguing over whether [p], [ph], and [b] are, or ought to be, distinct is about as fruitful as arguing whether Mary, marry, and merry are pronounced identically or entirely differently. It all depends: where are you from; how conscious are you of your own accent; to whom are you speaking; and how much do you care that your accent seem neutral to them?

  • 1
    This is a really interesting point and brings up how Indian English is unfairly marginalized in some discussions. Contextually it's worth remembering that we're talking about < 1M L1 speakers. However that is not a good reason to disregard the importance of Indian English in the development of the language. Commented Jun 3, 2023 at 15:09
  • @CynicallyNaive I’d dispute the < 1M figure. It counts only those whose language of family origin is English. It’s true that fewer that 1M Indians speak English with their families of origin. But far more than that, on the order of several million, speak English with native fluency simply because it’s the lingua franca. Bilingualism and multilingualism, where several languages are natively spoken, is a thing. Some U[SK] speakers, being monoglots, assume that the only true speakers of English are those who speak no other language fluently, which seriously distorts their linguistic analyses.
    – verbose
    Commented Jun 3, 2023 at 16:14
  • Dr Lindsey’s video is an example, where native == U[SK], and what’s more, only those with a very specific, normative accent.
    – verbose
    Commented Jun 3, 2023 at 16:14
  • 1
    It’s useful and interesting knowledge. It is, however, rare for an ESL speaker to intentionally adopt a South Asian accent.
    – Davislor
    Commented Jun 3, 2023 at 17:43
  • @Davislor that wasn't the point of my answer. The point is that the OP's question seems to assume that there is one "correct" accent, and that assumption seems unfounded. So what I'm saying to the OP is, don't fret about the correct way to pronounce those words; pronounce them in a way that works for you and your audience. What if the OP does come from a language background that maintains distinctions between [p], [ph], and [b]? Should they intentionally flatten those distinctions? I'd say, not unless they specifically wanted to do so for the sake of their audience.
    – verbose
    Commented Jun 3, 2023 at 18:27

Pinyin g and d are English unaspirated k and t.

one can put a hand or a lit candle in front of one's mouth, and say spin [spɪn] and then pin [pʰɪn]. One should either feel a puff of air or see a flicker of the candle flame with pin that one does not get with spin.

In English, stop is [stɑp], top is [tʰɑp], but day is [deɪ̯].

Pinyin d is [t], pinyin t is [tʰ], so dao is [taʊ] and tao is [tʰaʊ].

stamp should be pronounced as [stæmp] with an unaspirated t. That's not the same as English d (which normally represents IPA [d]).

The English d sound does not exist in Mandarin as far as I know. It is a voiced consonant, in English the vocal folds begin to vibrate before the consonant is released (voice-onset time).

In English, the stops p, k, and t are aspirated when they are at the start of a syllable, but not otherwise.

For example, port would be [pʰɔɹt], passport would still have aspiration [pʰæs.pʰɔɹt], but sport should be [spɔɹt] without aspiration.

So, you should only pronounce /t/ as [tʰ] if it is at the start of a syllable. It doesn't matter if it follows an s or not.


Should convert 'k' and 't' sounds to 'g' and 'd' sounds when they follow 's' in a word for pronunciation?

For example, stamp is pronounced as 's dæm p', etc.

How to answer this question depends on the context.

It's definitely true that the words stamp and skin are not pronounced with the same sounds that the letters T and K make in the words tank and kin. You are right to notice this.

You can use websites like Forvo and Youglish to hear how English speakers pronounce words like "stamp" (Forvo) ("stamp" Youglish) and "skin" (Forvo) ("skin" Youglish).

Imitating this as best as you can should result in an understandable pronunciation.

Why you hear a difference, and what that difference is

The phonetic difference is that words like tank and kin have a puff of air, called "aspiration", at the end of their T and K sounds, while "stamp" and "skin" do not.

This difference isn't represented in English spelling because English words can't start with an S sound followed by an aspirated T or K sound.

It's the same as the difference in Standard Chinese/Mandarin between the sounds written in pinyin with the letters t p k (which are aspirated) and the sounds written in pinyin with the letters d b g (which are not aspirated). So if you are familiar with that, and that's what you mean by writing 's dæm p', then that's a correct description of the start of English "stamp": the voiceless fricative S followed by an unaspirated voiceless consonant.

However, the letter d isn't generally used to represent an unaspirated voiceless consonant in the International Phonetic Alphabet. Instead, in this alphabet, we would mark the difference by transcribing "tank" and "kin" with [tʰ] and [kʰ], where the [ʰ] symbol represents the aspiration of the consonant, vs. "stamp" and "skin" with [st] and [sk] (with no [ʰ]).

The International Phonetic Alphabet prescribes using the letters [b d ɡ] to transcribe what are called "voiced" consonant sounds. These don't appear after word-initial [s] in English, but because they are also unaspirated, some English speakers hear a similarity between them and the voiceless unaspirated [p t k] sounds, as described by the-baby-is-you's answer. (The English phonemes transcribed /b d ɡ/ are not always fully phonetically voiced.) Other English speakers, or speakers of some other languages, may not hear this similarity, meaning for those people, it won't make sense to think of words like stamp, skin, spin as starting with sd-, sg-, sb- sounds. That's OK: while you can think of these words as starting with sd-, sd-, sb- if it helps you avoid using the aspirated consonant sounds found at the start of "tank", "kin", "pin", there's no requirement for everybody to think of them this way.

And of course, nobody should get the idea that they should pronounce these words with a phonetically voiced [b d ɡ] sound, so for speakers of languages where those sounds contrast with voiceless unaspirated [p t k], such as Italian, it makes the most sense to think of the English consonant clusters as being like their native [sp] [st] [sk].


P, T, and K aspirations are blocked by certain preceding sounds. See this video https://www.youtube.com/watch?v=U37hX8NPgjQ for how the sounds change and native English speakers don't even know what sounds we're actually making when speaking.

  • Your answer could be improved with additional supporting information. Please edit to add further details, such as citations or documentation, so that others can confirm that your answer is correct. You can find more information on how to write good answers in the help center.
    – Community Bot
    Commented Jun 4, 2023 at 8:05

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .