Audio Pairs system: Focus on Phonetics rather than Spelling

[DeeDubb]

Just saying that I've been using the "ng" sound instead of "h" at the end of letter pairs, and it's been working wonderfully.

That's probably the best discovery I made while writing my report. I'm loving it too. Thanks.

 

[newtonbase]

I've been using a few of these rules and I'm finding them really helpful. H = ng is my favourite too. Next is x = sh and q = th. I'm still playing around with them but hope to start a letter list soon.

 

[JemFish]

I've been using a few of these rules and I'm finding them really helpful. H = ng is my favourite too. Next is x = sh and q = th. I'm still playing around with them but hope to start a letter list soon.

Yeah me too. The [H = ng], [X = sh], [Q = ch], and [C = th] has been hard-wired into my brain so that now it's unconscious. If I hadn't read this post I would have still been under the Curse of English Phonetics in Letter Pairs.

 

[DeeDubb]

This was actually the best I was hoping for. I know old BLDers already have their set pairs that work, so it can't help them at this point, but I'm glad a few newer people like me are getting something out of this system

 

[cmhardw]

This was actually the best I was hoping for. I know old BLDers already have their set pairs that work, so it can't help them at this point, but I'm glad a few newer people like me are getting something out of this system

I think something like this is definitely a very good idea. If I were really practicing BLD like I used to I might consider switching to a more phonetic system like this. I think for people who aren't already entrenched in a system that learning a phonetic one like yours is a great idea. Thanks for putting forward so much leg work for others to improve with!

 

[TDM]

Just saying that I've been using the "ng" sound instead of "h" at the end of letter pairs, and it's been working wonderfully.

I've never see this before... This is such a good idea. H at the end of pairs is the worst thing ever for (normal) memo

 

[DeeDubb]

I think something like this is definitely a very good idea. If I were really practicing BLD like I used to I might consider switching to a more phonetic system like this. I think for people who aren't already entrenched in a system that learning a phonetic one like yours is a great idea. Thanks for putting forward so much leg work for others to improve with!

Thanks Chris, I really appreciate the feedback.

 

[AlphaSheep]

I'm still fairly new to blind, but I really struggle to remember images because I don't have much visual memory. I've found I have to rely on my audio memory to do blindfolded solving, because remembering sequences of sounds (even if they don't make proper words) seems to be much easier for me. On the rare occasions that a failure is due to bad recall, it's almost always because I've mixed up similar sounds like the ones you've highlighted.

I think a structured phonetics system like this is an perfect fit for someone who's brain works the way mine does, and may even be essential for when I eventually move on to multiblind and big cubes. The system will take a bit of practice to get used to, but what doesn't? Thanks for all your work.

 

[CHJ]

i understand what you've done here but i still think your audio is still too closely related to the actual words (i find linking to words makes me fail)
okso what i do for C and K (and S and Q), for example letter T, TK i'd use Tack, TC i'd use Tace so TK uses a hard consonant instead of the softer one in TC, then we come to the issue of TS and TQ, well TS is Tess for me and isn't extended as much as TC, Just as TQ is Teek and much more extended than TK.
When we get to H i think of it as an extended A, Ga would be Ga and GH would be Gaah.
also i do use pairs (so to say) but realistically i would say i use quads, so if i had a dodgy pair i'd influence its sound with the rest of that quad (6's if wings), i find if i cram it into a quad or a 6 then im less likely to forget it as if its associated with other audio pairs then i'd need to forget both pairs to forget it. sometimes i'm lucky enough to get 4 letter words which is a nice bonus.

 

[DeeDubb]

i understand what you've done here but i still think your audio is still too closely related to the actual words (i find linking to words makes me fail)
okso what i do for C and K (and S and Q), for example letter T, TK i'd use Tack, TC i'd use Tace so TK uses a hard consonant instead of the softer one in TC, then we come to the issue of TS and TQ, well TS is Tess for me and isn't extended as much as TC, Just as TQ is Teek and much more extended than TK.
When we get to H i think of it as an extended A, Ga would be Ga and GH would be Gaah.
also i do use pairs (so to say) but realistically i would say i use quads, so if i had a dodgy pair i'd influence its sound with the rest of that quad (6's if wings), i find if i cram it into a quad or a 6 then im less likely to forget it as if its associated with other audio pairs then i'd need to forget both pairs to forget it. sometimes i'm lucky enough to get 4 letter words which is a nice bonus.

I can't seem to remember unless they are actual words or something linked to an actual word.

 

[Stefan]

I can't seem to remember unless they are actual words or something linked to an actual word.

I believe you can make them "actual words". Just like when you learn a normal foreign language. "Ei" might not be an "actual word" for you right now, but you're surely capable of learning German, and then you'd have no problem using it as an alternative for "egg" and regarding it as an "actual word". If I ever get around to creating/learning letter pair words, I intend to make them all single-syllable words, most of which I'll probably have to invent. It'll just be my own little language.

 

[cmhardw]

Does anyone remember Macky talking about how an audio memo method is quite easy when using Japanese as the base language because of the consonant-vowel nature of Japanese? I believe he posted it on the yahoo group ages ago. Perhaps using a consonant-vowel pattern for your single syllables when possible, or even learning some Japanese may make audio memo even faster.

 

[suushiemaniac]

Does anyone remember Macky talking about how an audio memo method is quite easy when using Japanese as the base language because of the consonant-vowel nature of Japanese? I believe he posted it on the yahoo group ages ago. Perhaps using a consonant-vowel pattern for your single syllables when possible, or even learning some Japanese may make audio memo even faster.

Obviously I'm not Macky, but Japanese indeed makes things a lot simpler when using audio memo. Because unlike our Latin alphabet, which uses single letters, the Japanese alphabet consists of syllables called "kana". Nearly all of these kana are formed out of one (or sometimes two) consonants followed by a vowel which results in words like ma-gu-ro (tuna) or ka-ra-te.
When applied to BLD, this syllable system has the effect that any given audio loop can be directly pronounced. Instead of having to deal with CXGJFLMN (arbitrary example memo) where you have to use certain "tricks" to convert it to an audio line, you can get example memos like WA-MI-TSU-KE-TA-RU or NI-HO-N-GO-GA-HA-NA-SE-RU, fitting better into the natural human language flow.
I don't use a Japanese lettering scheme myself because I started off with Speffz and now I'm too lazy to switch. But given the advantages learning and/or using Japanese kana for BLD definitely is worth it if you have a strong audio memo.

 

[DeeDubb]

I created another system more similar to Japanese style but using English letter a while back. The problem is a consonant-vowel system means each paidr requires two syllables, and our consonant vowel consonant system can fit a pair into a single syllable. Doubling the length of the memo sequence makes retaining very difficult.

 

[irontwig]

Obviously I'm not Macky, but Japanese indeed makes things a lot simpler when using audio memo. Because unlike our Latin alphabet, which uses single letters, the Japanese alphabet consists of syllables called "kana". Nearly all of these kana are formed out of one (or sometimes two) consonants followed by a vowel which results in words like ma-gu-ro (tuna) or ka-ra-te.
When applied to BLD, this syllable system has the effect that any given audio loop can be directly pronounced. Instead of having to deal with CXGJFLMN (arbitrary example memo) where you have to use certain "tricks" to convert it to an audio line, you can get example memos like WA-MI-TSU-KE-TA-RU or NI-HO-N-GO-GA-HA-NA-SE-RU, fitting better into the natural human language flow.
I don't use a Japanese lettering scheme myself because I started off with Speffz and now I'm too lazy to switch. But given the advantages learning and/or using Japanese kana for BLD definitely is worth it if you have a strong audio memo.

How would being more phonetically restricted be better? Wouldn't it be better to be able to use one syllable for each 3-cycle (as you easily could do with any Germanic language) rather having one syllable for each target? Not to mention that you would be able to use actual words rather than a meaningless stream of syllables. Also, how the languages are written does not have anything to do with this at all.

 

[Stefan]

Nearly all of these kana are formed out of one (or sometimes two) consonants followed by a vowel which results in words like ma-gu-ro (tuna) or ka-ra-te.
[...]
memos like WA-MI-TSU-KE-TA-RU or NI-HO-N-GO-GA-HA-NA-SE-RU

Does one syllable encode one target or two? (I don't know how many vowels Japanese has and how many variations can be built)

 

[suushiemaniac]

Does one syllable encode one target or two? (I don't know how many vowels Japanese has and how many variations can be built)

One syllable = one target. There are 46 Japanese kana, enough to label a cube for BLD. You can even apply some rules of variaton to this basic stock of 46 to receive new combinations tending towards 100.

How would being more phonetically restricted be better? Wouldn't it be better to be able to use one syllable for each 3-cycle (as you easily could do with any Germanic language) rather having one syllable for each target? Not to mention that you would be able to use actual words rather than a meaningless stream of syllables. Also, how the languages are written does not have anything to do with this at all.

I personally don't see it as phonetic restriction. It's rather an entirely different system. I get your point and I don't think that our Indogermanic understanding/perception of sounds and syllables is especially bad. Using Japanese is simply another approach and in fact, the "meaningless stream of syllables" often arranges in an order that actually forms a Japanese word and sometimes even a complete phrase.

I created another system more similar to Japanese style but using English letter a while back. The problem is a consonant-vowel system means each paidr requires two syllables, and our consonant vowel consonant system can fit a pair into a single syllable. Doubling the length of the memo sequence makes retaining very difficult.

Using english letters is a nice idea, because you then don't have to speak any Japanese to use the system. But especially after reading irontwig's post I think that using Japanese syllables is only woth it if you have some basic knowledge of the Japanese language. Very often the memo either forms words (see above) or you can at least incorporate some basic linguistic patterns into your memo.
For example, Japanese is very regular when it comes to suffixes. Past tense verbs all (correct me if I'm wrong) end in either "TA" or "DA". So whenever one of those syllables occur in your memo, you can imagine that something happened in the past. I know that imagination and mental depicting doesn't apply to audio memo, but at least to a lot of other memo systems.

 

[irontwig]

I personally don't see it as phonetic restriction. It's rather an entirely different system.

It's not an "entirely different system". Sure there's morae and all that, but e.g. tonal languages differ far more in phonotactics (albeit not all that much even there). It sounds more like you're starring yourself blind on the writing system.

 

[tseitsei]

One syllable = one target. There are 46 Japanese kana, enough to label a cube for BLD. You can even apply some rules of variaton to this basic stock of 46 to receive new combinations tending towards 100.

If 1 syllable encodes only 1 target then it is clearly not as good as "normal" (commonly used) phonetics system where 1syllable encodes 2 targets. You'll have to memo twice the amount of syllables to memo the same amount of information. Obviously not good and very inefficient...

 

[DeeDubb]

It's not a matter of knowing English. It's the amount of information encoded per syllable. Japanese use their system because it's what their language allows. Restricting ourselves to the same thing doesn't make sense.