• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 30,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

Audio Pairs system: Focus on Phonetics rather than Spelling

Joined
Mar 9, 2014
Messages
1,432
Likes
30
Location
South Korea
WCA
2014WHIT07
YouTube
DeeDubbCubing
Thread starter #1
I'm working on my letter pair spreadsheet, which you can see here.

There are some things I am still working out, but the main thing is I'm working on consistent pronunciation. There's some issues with English letters not matching English pronunciation, so I'm trying to make my spreadsheet as consistent as possible while still finding ways to make sure there is solid association.

One thing that is important: Dialects and pronunciations can vary. My pronunciation is a fairly neutral American accent, but I believe that my system should fit fairly well with most dialects of English. Please let me know if this is not the case for other speakers of English.

Difficulties:

For 3BLD, we often label the corner/edge stickers A-X, and for letter pairs, we need words that fit from AA - XX. This is 24 letters. The following are problems with fitting some of these letter pairs to words:

Consonants:
- C and K generally sound the same.
- Q has very limited environments and often sounds like C/K as well (especially in syllable-final position)
- X has very limited environments, and usually only exists at the end of a word. Plus, the pronunciation is similar to "ks". At the beginning of a word it will sound like a "z" (this isn't a huge problem because we don't need z for our A-X labeling.
- H is only pronounced at the beginning of syllables, and not at the end. You hear the "h" in "hey" but not in "yeah".

Vowels:
- Vowel pronunciation is VERY inconsistent in English. This is because of multiple language influences as well as our language having way too many vowel sounds (around 16) for our inventory of only 5 vowels (a, e, i, o, u).
-Making words for AO, AU, EO, etc... These are very tricky to do phonetically.

So, how can we deal with these? If we assign phonetic sounds to each letter, then we open up a wider variety of words as well as create more consistency to our audio pairs and easier recall (because we can recall based on actual sounds rather than an arbitrary letter).

Before we can start assigning, we need to understand all of the sounds that English commonly has in order to figure out where we/how we can systematically and efficiently assign these

Vowels:
note: vowels are most affected by dialect, so there may be some variation. I'm using my own personal dialect for this list
/i/ - "beat"
/I/ - "bit"
/e/ - "bait"
/ɛ/ - "bet"
/æ/ - "bat"
/ɑ/ /ɔ/ - "bot" "bought" (some people might pronounce these differently, I don't)
/ʌ/ /ə/ - "but"
/o/ - "boat"
/ʊ/ - "book"
/u/ - "boot"

glides:
/aI/ - "bite"
/ɔI/ - "boy"
/aʊ/ - "bow"


Consonants:
/p/ - standard "p" sound. Voiceless bilabial stop. Can be used at the beginning or end of a syllable (e.g. "pot" "top")
/b/ - standard "b" sounds. Voiced bilabial stop. Can be used at the beginning or end of syllable (e.g. "bored" "web")
/t/ - standard "t" sound. Voiceless alveolar stop. Can be used at the beginning or end of a syllable (e.g. "tame", "bet")
/d/ - standard "d" sound. Voiced alveolar stop. Can be used at the beginning or end of a syllable (e.g. "Deal", "Bad")
/tʃ/ - t + "sh" sound, written as "ch" in English. Can be used at the beginning or end of a syllable (e.g. "Chair", "Batch")
/dʒ/ - /d/ + post alveolar fricative /ʒ/ (sounds like the "g" in "beige"). Written as "j" or "dg" in English. Can be used at the beginning or end of a syllable (e.g. Jam, Budge)
/k/ - standard "k" sound. Voiceless velar stop. Usually written as "k", "c", or "ck" in English. Can be used at the beginning or end of a syllable (e.g. "care" "tick")
/g/ - standard "g" sound. Voiced velar stop. Written as "g" in English. Can be used at the beginning or end of a syllable. (e.g. "gear", "beg")
/f/ - standard "f" sound. Voiceless bilabial fricative. Usually written as "f" or "gh" in English. Can be used at the beginning or end of a syllable (e.g. "fail", "rough")
/v/ - standard "v" sound. Voiced bilabial fricative. Usually written as "v". Can be used at the beginning or end of syllables (e.g. "vile", "cave")
/θ/ - First of two "th" sounds in English. Voiceless labiodental fricative. Can be used at the beginning or end of syllables (e.g. "Thin", "bath")
/ð/ - Second of two "th" sounds in English. Voiced labiodental fricative. Can be used at the beginning or end of syllables (far less common than the first "th" sound). (e.g. "then", "bathe").
/s/ - standard "s" sound. Voiceless alveolar fricative. Can be used at the beginning or end of syllables (e.g. "sail", "toss")
/z/ - standard "z" sound. Voiced alveolar fricative. Usually written as "z", often written as "s" at the end of syllables (e.g. "zap", "rose")
/ʃ/ - standard "sh" sound. Voiceless postalveolar fricative. Usually written as "sh" in English. (e.g. "shell", "mash")
/ʒ/ - Very rare sound in English. Mostly from French borrowed words such as "beige". We won't worry about this one too much
/h/ - standard "h" sound in English. Glottal fricative, only used at the beginning of syllables (sometimes written at the end of words such as "yeah", but the "h" isn't pronounced). (e.g. "heal", "who")
/m/ - standard "m" sound in English. Bilabial Nasal. Usually written as "m" and rarely as "mb" at the end of words, used at the beginning or end of syllables. (e.g. "mail", "lamb").
/n/ - standard "n" sound in English. Alveolar Nasal. Usually written as "n" in English. Written at the beginning or end of syllables ("Name" "Man")
/ŋ/ - "ng" sound in English. Velar Nasal. written as "ng" in English. Though it is written as "ng", it is a singular sound rather than a combination of "n" and "g". Only exists at the end of syllables in English words. (e.g. "thing")
/l/ - standard "l" sound in English. Lateral Approximant. Usually written as "l". Can be at the beginning or end of a syllable. (e.g. "Like", "Bell")
/ɹ/ - standard "r" sound in English. Approximant. Usually written as "r". Can be used at the beginning or end of a syllable. (e.g. "Rub", "Bar")
/w/ - Semivowel. Voiced Labio-velar approximant. Written as "w". Only exists at the beginning of words. Written at the end often, but only pronounced as a vowel at the end. "Now" is phonetically /naʊ/, and doesn't contain a "w" sound. (e.g. "what", "wood")
/j/ - Semivowel, palatal approximate. written as "y". Only exists at the beginning of words. Written at the end often, but only pronounced as a vowel sound. (e.g. "Yes", "yield")


As you can see, we have a wide inventory of sounds, but the pronunciation doesn't match the writing often. For audio pairs, try to ignore the way the word is written and focus on how it sounds.

Among our 24 letters, the following have very clear letter assignments:

b - /b/
d - /d/
f - /f/ (also remember words like "rough" would go here)
g - /g/ (remember "ng" words do NOT go here, because it's not a "g" sound)
l - /l/
m - /m/
n - /n/
p - /p/
r - /ɹ/
s - /s/
t - /t/
v - /v/


The following letters can easily be assigned a sound:

j - /dʒ/ (remember, this will apply to a "j" at the beginning of a syllable or a "dge" at the end, such as judge)

k - /k/ (remember, "c" words will ALSO go here, if they sound like /k/, so "cat" would fall under KT rather than "ct")


We can fill in some more by using some extra sounds (ones without a single letter associated to them):

c - /tʃ/ aka the "ch" sound. using "ch" for all "c" pairs clearly differentiates it from k while still being intuitive enough to clearly remember. "ch" is prevalent both at the beginning and end of syllables, so this is not a problematic assignment.

x - /ʃ/ (the "sh" sound. This makes intuitive sense if you consider the Chinese names with "x" at the beginning sound like "sh").
- /ks/ ("ks" also works, but only at the end of words, such as "Tux" for TX or "socks" for SX. Basically find the best "sh" or "ks/x" word ending, and choose that)

h - /h/ works fine at the beginning of syllables, but fails at the end. We can solve this by either assigning a specific vowel to /h/ for the end of syllables or finding a sound that only exists at the end of syllables.
- /ŋ/ Fortunately, we have the /ŋ/ (aka the "ng" sound)which only exists at the end. Meaning, any time we have H(X) pairing, we'll use /h/ sound, and any time we have (x)H pairing, we should use /ŋ/ instead.

q - /kw/ feel free to use the traditional q sound for some words. Such as "Quick" for /KW/ However, when you run into problems...
- /θ/ and /ð/ (The "th" sounds. There's no really intuitive connection, but after "ch" and "sh" sounds, "th" is most common leftover sound) This works well at the beginning and end of syllables, so feel free to use /kw/ or /th/ at the beginning of syllables, while you'll be using only /th/ at the end.

w - (see below) This one is tough. /w/ works fine at the beginning of syllables, but is more problematic at the end, (for example, the word "Chow" would be tough to know whether it's /CW/ or /CO/. I will assign w with my vowels to prevent confusion. Also we have a very large vowel inventory so assigning extra sounds to /w/ should not be a problem.

Vowels:
Remember: We are focusing on the sounds rather than the letters... "Bee" and "Tea" have the same /i/ vowel sound, so they are under "i" rather than "e" or "a"

a - /æ/ /ɑ/ /ɔ/
Examples:
AB - "ab" /æ/
AF - "off" /ɑ/ /ɔ/


e - /ɛ/ /e/ /aI/
ET - "ate" /aI/
TE - "tray" /e/
EX - "ex" /ɛ/

i - /i/ /I/
IL - "eel" /i/
IF - "if" /I/

o - /o/ /ɔI/
XO - "show" /o/
OL - "oil" /ɔI/

u - /u/ /U/
XU - "Shoe" /u/
RU - "Roux" /u/
UP - "up" /ʌ/ /ə/

w - /w/ (beginning) /aʊ/ (end)
WL - Wheel /w/
LW - Lau (Alex!) /aʊ/
BW - Bow (like 'Bow down") /aʊ/

Leftover sounds:
/z/ - you can use it if you have trouble finding a pair for S, but not super necessary.
/ʒ/ - Too rare to be convenient. I would not use this sound.
/j/ - Not really needed. Could be used for some tricky vowel on vowel pairs (i.e., IA could be "Yeah", IU could be "You", IO could be "Yo")



Remaining difficulties:
Finding Vowel/Vowel combos can still be problematic. I think I will just have to memorize associations for each of these.
Getting accustomed to how the vowels work could be difficult. For me, it's very easy because of my background in Linguistics. However, I believe anyone could grow to understand this with too much difficulty.
 
Last edited:
Joined
Jan 27, 2015
Messages
338
Likes
0
Location
Sydney, Australia
#2
I haven't read the whole thing, but I will soon. I've been having some problems with audio pairs, and hopefully this will help. Thanks for the effort you put into this post.

EDIT: Hey, this is very good. I really like how your post covers British/Australian/Kiwi/Indian/Singaporean etc. accents; I'm also really glad that I know how to read IPA (International Phonetic Alphabet).

I also like how you have the /ʃ/ sound for X, similar to Chinese, and so a suggestion would be to have the /tʃ/ sound for Q, also similar to Chinese, and giving the /θ/ and /ð/ sounds to C instead.
 
Last edited:
Joined
Mar 9, 2014
Messages
1,432
Likes
30
Location
South Korea
WCA
2014WHIT07
YouTube
DeeDubbCubing
Thread starter #3

I also like how you have the /ʃ/ sound for X, similar to Chinese, and so a suggestion would be to have the /tʃ/ sound for Q, also similar to Chinese, and giving the /θ/ and /ð/ sounds to C instead.
Thanks for the feedback. That's a fine suggestion. I won't change mine, because "ch" has a more intuitive connection to "c" than "q" for me, but I can certainly see your point, and you should go with it for your audio pairs list.
 
Joined
Apr 23, 2010
Messages
1,382
Likes
136
Location
Scotland, UK
WCA
2009SHEE01
YouTube
bobthegiraffemonkey
#4
I do something like this for my audio loop (without using pairs, just sounds, but its very little information so it's fine) except you obviously know far more about this than I do. As for tricky vowel/vowel combos, I set aside a consonant which I can throw in between the vowels to make it nicer, and vice versa for tricky consonant/consonant combos.
 
Joined
May 15, 2014
Messages
37
Likes
1
WCA
2014HALE02
#5
It's funny how I had exactly the same idea a while ago.
In the end, what I came up with was almost identical yours, since people's association of how each letter should sound doesn't differ by much.
 
Joined
Mar 9, 2014
Messages
1,432
Likes
30
Location
South Korea
WCA
2014WHIT07
YouTube
DeeDubbCubing
Thread starter #6
It's funny how I had exactly the same idea a while ago.
In the end, what I came up with was almost identical yours, since people's association of how each letter should sound doesn't differ by much.
Yeah, I think it's fairly intuitive. I just wanted to explain each element in a systematic and linguistic way.

Wait, you're Korean? Are you in the 큐브 facebook group?
 
Joined
May 15, 2014
Messages
37
Likes
1
WCA
2014HALE02
#8
Yeah, I think it's fairly intuitive. I just wanted to explain each element in a systematic and linguistic way.

Wait, you're Korean? Are you in the 큐브 facebook group?
Indeed, I am. However, Korean isn't even my best language. I live in Singapore; one might be able to infer it from looking at which competitions I have gone to.
I certainly wouldn't mind being in the Facebook group. In fact, I'd love to be in it. But I think it'll be even more awkward when a native that can't even type his own language properly is present. Haha.
I'll PM you my FB profile later so you can add me. Is that alright?
 
Joined
Mar 9, 2014
Messages
1,432
Likes
30
Location
South Korea
WCA
2014WHIT07
YouTube
DeeDubbCubing
Thread starter #9
Indeed, I am. However, Korean isn't even my best language. I live in Singapore; one might be able to infer it from looking at which competitions I have gone to.
I certainly wouldn't mind being in the Facebook group. In fact, I'd love to be in it. But I think it'll be even more awkward when a native that can't even type his own language properly is present. Haha.
I'll PM you my FB profile later so you can add me. Is that alright?
Yeah, I saw your profile, and obviously your English is great. That's why I asked if you were in the Facebook group rather than if you lived in Korea ;) Sure, PM me, and I'll add you.
 
Joined
Jan 12, 2012
Messages
1,361
Likes
45
Location
Tampere, Finland
WCA
2012LEHT01
#10
Great thing in finish language for audio loops is that you pronounce everything the same way you write them so making audio loops is very simple :tu
The problem in finish is that we don't really have words that use C/D/B that much so I just use english words for those...
 
Joined
Mar 9, 2014
Messages
1,432
Likes
30
Location
South Korea
WCA
2014WHIT07
YouTube
DeeDubbCubing
Thread starter #11
Great thing in finish language for audio loops is that you pronounce everything the same way you write them so making audio loops is very simple :tu
The problem in finish is that we don't really have words that use C/D/B that much so I just use english words for those...
Wow, interesting! I'm guessing that most words with that "c" sound we use are spelled with a "k" in Finnish? Are P and T very common? Because those are the voiceless alternatives to D and B. I don't know anything about Finnish, but I haven't seen any languages that take out voiced alveolar /d/ and voiced bilabial /b/, as those are some of the most common consonants across all languages.
 
Joined
Sep 5, 2014
Messages
275
Likes
4
Location
Tennessee
#12
Awesome job. I used to be really into linguistics, so it's cool to see a fellow cuber who is knowledgeable about it. I don't use ɔ in my dialect, but I think of it as being closer to o than a. Like you said, it depends on the individual to customize it to their preferences. I actually already have a way to read letters phonetically that I use to remember license plates and random strings of letters even though I don't do BLD. Nice work.
 
Joined
Jan 12, 2012
Messages
1,361
Likes
45
Location
Tampere, Finland
WCA
2012LEHT01
#14
Wow, interesting! I'm guessing that most words with that "c" sound we use are spelled with a "k" in Finnish? Are P and T very common? Because those are the voiceless alternatives to D and B. I don't know anything about Finnish, but I haven't seen any languages that take out voiced alveolar /d/ and voiced bilabial /b/, as those are some of the most common consonants across all languages.
yeah P,K and T are all very commonly used. And yes we use K (almost) always instead of C
 
Joined
Mar 9, 2014
Messages
1,432
Likes
30
Location
South Korea
WCA
2014WHIT07
YouTube
DeeDubbCubing
Thread starter #16
Spanish is very consistent with regard to spelling and pronunciation. I may base my words around that. Especially the vowels. I don't speak it much but am familiar enough.
yes, I tried to use latin group vowels, for a, e, i, o, u. However, we have such a wide inventory of vowels in English, it would be very limiting to only use those basic 5 sounds for our letter pairs.
 

newtonbase

Premium Member
Joined
Jan 4, 2014
Messages
2,940
Likes
2,669
Location
MK, UK
WCA
2015ADAM03
YouTube
newtonbase
#17
I was thinking of not necessarily using real words. I'd only be doing it for corners so around 5 distinct one or 2 syllable words remembered purely as sounds shouldn't be hard to remember for a few mins.
 
Joined
Mar 9, 2014
Messages
1,432
Likes
30
Location
South Korea
WCA
2014WHIT07
YouTube
DeeDubbCubing
Thread starter #18
I was thinking of not necessarily using real words. I'd only be doing it for corners so around 5 distinct one or 2 syllable words remembered purely as sounds shouldn't be hard to remember for a few mins.
For me, I've noticed I have a much harder time remembering arbitrary sounds as opposed to real words, even for audio pairs. Also, I use my audio for edges, so it could be 6-7.

Plus, to be honest, my goal eventually is to do the entire cube with audio pairs, so they have to have some meaning to retain.
 

newtonbase

Premium Member
Joined
Jan 4, 2014
Messages
2,940
Likes
2,669
Location
MK, UK
WCA
2015ADAM03
YouTube
newtonbase
#19
That makes sense. I expect full audio pairs would be quicker to memo than a journey so having real words would help. I'm still working on getting a decent strike rate on corners before I move on to edges.
 
Top