Appendix talk:Welsh pronunciation

About accented letters
What is in Cymraeg, the phonetic of the letter "ë", "Ë"?

Is-it "é [e]", like in "bed"

"i [ee or i:]", like in "bee"

"ai [ɛ or ei], like in "aid"?

But the Welsh is closer Old French language = Old Loegrian (= Early French from Liger, Loire, Long-Rhyn, Ligurian, Lydao, LLydaw = all north of France and New-LLoegr = the Two-faced-England), than the Saxon-English. So i am inclined to say that "ë,Ë" sound like in "Noël, Noë [nɔe]" in "FFrainc" = "é [e]" like in "Set". This sound doesn't exist really so clearly, in English-saxon; isn't-it?

I notice that for the "Ŵ,ŵ", the sound is not only the "OU, u:" like in "boo"; but also the diphthong sound "OA", "A-in-O, like in "boat [bəʊt]" or in Welsh, like in "gŵaraig" (a woman, female); it's also "OI" like in "oily [ˈɔɪlɪ]" in other cases.

Generally only one letter has always to the minimum of 4 manners to be say; & with all the local accents much much more, of course.

Re: ë This does not denote a specific sound in Welsh - it is a diaeresis, indicating that the e is pronounced separately from the vowel which precedes it (i.e. it does not form a diphthong). It is used in the same way in English in the names Chloë nd Zoë. In general, e in Welsh is similar to English short e (as in 'get'), whilst ê is similar to French é, but longer.

LL equivalent
You said : "llyfr, cyllell No English equivalent...", yes you have in WeLSH [welʃ] mutation of GALL [velʃ]. It is an equivalent; that means it's not like the original, but nearly similar.

LL sound in LSH, SHL (Same if now in English LL=FL, like in LLOYD = FLOYD); but like in French, the word "CHLEU, CHLEUH [ʃlø]" that we could write LLEU [ʃlø]; in FRENCH "FRENLL" [frɛnʃ], final "ELCH, ELSH [elʃ] = LL". When we talk about "equivalent" is not to must have the same, but the more similar that we can find.

All will be easier and more simple, if we could listen some audio samples; as do the chineses online dictionaries. With a real audio sound, no risk, no chance, to make errors. Please! if you want to propagate the Welsh language, the Welsh have to do the same (I address this message to the Universities of Wales, more busy to find from "Celt" sick of celtomanic, from celtomaniac, true falsifiers and fakers).


 * If anything, I'd say the "thl" in athlete is the closest thing we have in English to the Welsh [ɬ]. But you're overestimating the value of audio files if you think they result in "no risk, no chance, to make errors". Mimicking back an unfamiliar sound once you've heard it (especially through loudspeakers rather than in person) is not really a reliable way to learn the correct pronunciation of a foreign language. —Angr 17:38, 17 February 2013 (UTC)

Perhaps? But it's wrong. We are a lot of people to be unable to speak with real Welsh people, same if we would like to do it. I listen often BBC-Cymru and I use my French-ears (like a cute & clever frog), the words with 'LL' sound in 'SHL' [ʃl] or 'LCH [lʃ]. Do you think my ears translate the sounds in [FL]? No of course. Yes it's a good manner to learn a tongue with audio files. All the people make that to learn and begin to speak in a foreign language. So audio files are indispensables or essential to repeat right sounds of a foreign tongue.


 * Angr, I suppose it depends what you mean by "in English". I'm not sure whether the combination "thl" occurs in any English words. Personally I don't think that gets very close to "ll". Many English speakers seem to go for "khl" [xl], which is not bad. But then in my experience most English speakers don't find "ll" all that difficult anyway. It's other things, like diphthongs, that give much more trouble than this supposedly unpronounceable sound. And while of course it's true that [ɬ] doesn't exist in English as a phoneme, we do hear a trace of it in casual pronunciations of English words like "p(o)lice" ("plleece" [pɬi:s]) and emphatic "please!!".
 * However, Anon OP, "ll" [ɬ] is not in any sense I can see the "equivalent" of English or French "sh/ch" [ʃ], and I agree with Angr that sound files on a computer are not a reliable way of learning an unfamiliar speech sound, as perhaps is demonstrated by Cleverfrog's impression that it sounds like "shl" or "lsh" -- it really doesn't, in real life. Flapdragon (talk) 18:22, 17 February 2013 (UTC)
 * (Of course I was wrong to say "thl" doesn't happen in English, we have "athlete" and so on. I suppose I meant that "thl" isn't found as a single sound in English, only across a syllable boundary. It's never initial or final.) Flapdragon (talk) 18:57, 17 February 2013 (UTC)


 * Sure, we are behind our computers, may be it's not like in the real life, but it's a life real (It's a false debate). I ask me why, some companies make audio-CD, audio-translators, to allow people to start right the learning of a foreign tongue, if all that will be so unreliable? May be you prefer to go in a concert, full of noise, to appreciate music. I prefer listen quietly music in my living room. Could you accept the same, to start on good bases, with a special manner of some language, to say the right sounds of the letters; language is only music.
 * Yes 'KHL' could to be also good, because K=S, C, CH, SH; but 'K' is also very often 'YR, AR, RA' in East Welsh tongues (from Eurasian), and 'K' [ka, ca] in French is like 'Q' [ky, kə]; so from my viewpoint 'K' is not so good in the real life (lol).
 * Sound files are very helpful – they're definitely better than nothing! – but it's an exaggeration to say that having a sound file produces "no risk, no chance, to make errors". Even hearing the sound in person doesn't guarantee that. When I was learning Welsh, merely hearing the [ɬ] wasn't enough, the teacher had to tell me explicitly exactly what to do with my tongue in order to produce it. And when I was learning Lower Sorbian I never did learn whether the sound spelled w and ł (which represent the same sound in that language, unlike Polish) is pronounced [w], [v], [ʋ], or [β] – the sound I heard both from my (native-speaker) teacher and in the audio files I have sounds sometimes like one of those, sometimes like another of them. —Angr 21:27, 17 February 2013 (UTC)
 * Oh lala! please, don't be focussed on my 3 words "no risk, no chance & no errors", it's just a manner to speak; a insistence form. Of course there are always many differences to hear langages, and more accents with differents dialects inside a same country... Look the same cockcrow (rooster call) everywhere in world. French "Cocorico", English "cock-a-doodle-doo". Who has never heard the right sound of this big stud-chicken? (French [frɛnʃ] and not [frentʃ] as said Anglo-saxon of LLoegr [lwar'], is analogic-tongue, a tongue of Poët [pɔ-ɛt'] not said like [pəʊɪt]. So, all, we heard the same thing, but all, we write differently for the confort of our ears. Have you note when a Chinese spells "A ROOM", he says always "A LOOM", like the Ancients Iapitean (Eurasian); R=L and L=R; always in 21th century.
 * Now, if I said that the 'LL' = LSH, LCH [ɛlʃ] or SHL, CHL [ʃlə], it's because the letters used well before the middle-ages, to script the Welsh 'LL', are these letters; and I don't write all variations to pronounce the "C, S, SH, CH". Anyway, 'LL' is also often converted in French with a only one 'L' (like a Long-L). Example: The GALLES [gæl: or vɛlʃ] = WELSH [welʃ], Cymmrieg, Gallois [gɑ:lwɑ:, gælwɑ:, or 'ɑ:lɔɪ, &c] = VELCHES [vɛlʃ] (the Helvetian, French-Swiss); OLL GALLI [gali or vɛlʃ]; ALL WELSH since more 4500 years BC. The 'G' letter, often elided (mute, silent) has also several sounds, and takes differents letters as 'V, U, H, Y, C (S), CH, K, GU, &c'. So, with audio-files we can hear all that, clearly. After, what you do between your ears, it's an "other pairs of sleeves" ("different kettle of fish", said English of LLoegr, may be he thinks to the "Channel" (calls "Manche = Sleeve" in France), and not to a "item of garment". Have we the same brain? Never.

Is there another native Welsh speaker who can back me up when I say that ll is absolutely nothing like welsh or athlete. khl is about as good as you'll get from an English speaker, but it's still a little further forward in the throat than is correct. cwbr77 10:16, 10 January 2014 (UTC)
 * "Absolutely nothing like" is an exaggeration. The /θl/ in athlete isn't the Welsh ll sound, but it's not that far off, especially for English speakers who can't produce a "kh" to begin with. "Thl" is certainly a lot closer to the Welsh ll sound than, say, /b/ or a vowel sound would be! 11:26, 10 January 2014 (UTC)
 * I stand by it, the /θl/ sound is made with the tongue at the top of the mouth and no closing of the throat, whereas the LL sound does not involve the tongue and has the throat narrowed. Now that you mention vowels, the shape of the mouth for the "e" in athlete is quite similar, but the throat is a little tighter, so as to make a noise, and it's unvoiced. cwbr77 (talk) 17:33, 10 January 2014 (UTC)
 * The LL sound does not involve the tongue??? Then what's that thing pressed up against the alveolar ridge when you make it? And the throat isn't closed (which would make the sound glottalized) but open, since it's voiceless. Anyway, I've added a link to the Wikipedia article and a sound file so people can get a better idea of how the sound is made and what it sounds like. —Aɴɢʀ (talk) 17:53, 10 January 2014 (UTC)
 * Nothing goes anywhere near the alveolar ridge (which I did look up to make sure) in making the LL sound, which is done entirely in the back of the throat. All the Welsh teachers at my school taught it that way, and I spent years listening to them do it that way. I did it that way on my oral exam and had no complaint about my enunciation. I have had a loot at another, clearer explanation of that method of attempting to make the sound (here) and it does sort of work, but it seems like a very complicated way of getting the right sound. Even so, the explanation that it is like the voiceless "l" in please is nowhere near adequate and I propose that it is at least expanded on. cwbr77 (talk) 23:56, 10 January 2014 (UTC)
 * According to, some speakers use the instead of the . Your description, however, sounds sort of like the  ascribed to Welsh ch. It could be my personal subjective impression, but  seems to go with uvular articulation in some cases, which would explain your reference to "closing of the throat". It doesn't seem to be standard for Welsh, though. It's not that uncommon for individuals to substitute sounds in their own speech: my mom used to tutor someone who used to say [θ] instead of [f]. Chuck Entz (talk) 02:10, 11 January 2014 (UTC)
 * That seems quite accurate. I have attempted the 'proper' way of pronouncing words like "llygoden" and it would certainly explain why "ll" and "ch" are different letters, and why the double L is used for "ll". I wonder if perhaps the is taught to people whose first language is English due to it being easier to grasp. Certainly I never noticed that the sounds were supposedly different, even from something as deeply Welsh as Pobol y Cwm. As for the [θ] instead of [f], most of my partner's family do that, but the other way around. Thank you for your input, Chuck Entz, it's been very enlightening. cwbr77 (talk) 20:06, 11 January 2014 (UTC)

Fortunately, the pronunciation of Welsh words is regular and phonetic.
Since the pronunciation of words is always phonetic in any spoken language, what is this statement supposed to mean?


 * I suppose it means there's a grapheme-phoneme correspondence ("if you can spell it, you can say it"). Equinox ◑ 19:53, 4 September 2016 (UTC)
 * Which is closer to being true for Welsh than for English or the Goidelic languages, but isn't 100% true for Welsh either. —Aɴɢʀ (talk) 14:18, 5 September 2016 (UTC)

Long diphthongs
Could anyone confirm the presence of long diphthongs, as in the table, for example /aːu̯/ vs. /au̯/ ? Is there really such a phonemic contrast? It doesn't seem to be mentioned in Hannahs' The Phonology of Welsh (section 2.3.2) or the 2001 article on Northern Welsh in the Journal of the International Phonetic Association, or in the Wikipedia article. It does seem to be reflected in Wiktionary entries. My best guess is that this is related to the length variation mentioned in the next section of the book (2.3.3), but in this case it would seem to be not phonemic. Craig Baker (talk) 16:33, 22 December 2023 (UTC)


 * I don't think it's phonemic. —Mahāgaja · talk 21:25, 22 December 2023 (UTC)