Return-Path: <nifl-assessment@literacy.nifl.gov> Received: from literacy (localhost [127.0.0.1]) by literacy.nifl.gov (8.10.2/8.10.2) with SMTP id g6IDo8X04014; Thu, 18 Jul 2002 09:50:08 -0400 (EDT) Date: Thu, 18 Jul 2002 09:50:08 -0400 (EDT) Message-Id: <20020718134835.43386.qmail@web14308.mail.yahoo.com> Errors-To: listowner@literacy.nifl.gov Reply-To: nifl-assessment@literacy.nifl.gov Originator: nifl-assessment@literacy.nifl.gov Sender: nifl-assessment@literacy.nifl.gov Precedence: bulk From: tom zurinskas <tzurinskas@yahoo.com> To: Multiple recipients of list <nifl-assessment@literacy.nifl.gov> Subject: [NIFL-ASSESSMENT:157] R&D in phonetic awareness X-Listprocessor-Version: 6.0c -- ListProcessor by Anastasios Kotsikonas Content-Type: text/plain; charset=us-ascii Status: O Content-Length: 4981 Lines: 188 For adult literacy, I think there is a real opportunity in the ESL area with learners that are literate to advance phonetic awareness methodology and application. With truespel (which has respelled English in USA accent), the learner is exposed to an intermediary English friendly phonetic spelling. The sound associations of the phoneme spellings can be learned in a few minutes (only the vowels give trouble). Training would be done by an instructor in learners 1st language. Training "locks in" the association between sound and spelling of 40 English phonemes, which takes only an hour or two of practice. Nonsense words are used in practice, which are language neutral and are balanced in presentation for number of phonemes. After an hour or two, these already literate learners should be able to read aloud truespeld English text with maximal USA accent. They won't know what it means, but the decoding part is simplified. They will also be able to encode/decode their first language (given a few new phoneme spellings) which will also enforce learning the basic truespel phoneme set. Access to truespel is no problem. It is available for free via the truespel converter, which enables any teacher or learner to easily convert English to truespel. In the future truespel will be an intermediary phonetic spelling between all languages. This is possible because I've found that the 40 phoneme set of English makes up about 95% of the phonemes of other most-popular languages (for 13 investigated). New research is presented below, enabled by the truespel database. I'll be glad to support any research in this area. A 1-hr CD is now available to learn truespel (15 min), practice (35 min), and test yourself on your ability to hear phonemes (10 min). This is a new product and your support would be great. Analysis of English in USA Accent Developing a Truespel Leengwuprint Tom Zurinskas 7/4/02 I've completed the analysis of the English phonemes as appearing in the London Times (USA accent). The sample is made up of words appearing 100 times or more, or 10,281 individual words out of a 16,324,176-word sample. Here is a list of all words over 1% of the top 10,281 words. (The word “the” is far and away most popular.) The procedure consists of converting the words to truespel, which is a phonetic spelling, separating the phonemes associated with frequency, and tallying the results per phoneme with % computed. The second and third appearances of phonemes in words make up 6.2% of the total. (There were no words with 4 appearances of the same phoneme.) Thus, if the London Times is written as English is commonly spoken, then this is a fair representation of the frequency of phonemes in spoken USA English. Word Frequency Percentage the 1,081,654 6.63% of 535,391 3.28% and 511,333 3.13% to 479,191 2.94% a 419,798 2.57% in 334,183 2.05% that 215,322 1.32% it 198,578 1.22% i 197,055 1.21% was 194,286 1.19% is 166,691 1.02% Total 16,324,176 The data below I call it a leengwuprint. I display it also in a graph, phonemes along the x axis and percent utilization on the vertical axis. Other languages are compared to USA English visually this way with their own leengwuprints. Truespel Analysis of USA English phoneme usage from words in the London Times Newspaper total sample word count = 16,324,276 total individual words = 10,281 total phonemes = 62,053 avg. phonemes/word = 6.04 7/4/02 Vowels neem freq. sample v1 i 7.01% in v2 u 6.76% up v3 ee 4.46% meet v4 ie 3.97% pie v5 a 3.45% ad v6 e 2.48% end v7 er 2.34% her v8 ue 2.05% blue v9 aa 1.55% baa v10 ae 1.51% sundae v11 oe 1.19% toe v12 oo 1.04% good v13 or 0.68% fork v14 au 0.63% auger v15 ou 0.58% out v16 air 0.54% fair v17 oi 0.10% oil 40.32% Cons. neem freq. sample k1 n 7.94% nap k2 t 7.06% tap k3 s 4.50% sad k4 d 4.47% did k5 l 3.84% lip k6 th 3.18% that k7 z 2.97% zap k8 k 2.95% kid k9 r 2.84% ran k10 m 2.76% man k11 w 2.41% woe k12 v 2.05% vest k13 p 1.98% pad k14 g 1.81% gift k15 b 1.78% big k16 f 1.73% fig k17 h 1.68% hoe k18 y 1.13% yes k19 sh 0.77% sham k20 thh 0.66% thin k21 ch 0.59% chin k22 j 0.51% jig k23 zh 0.05% vision 59.68% all phonemes neem freq. 1 n 7.94% 2 t 7.06% 3 i 7.01% 4 u 6.76% 5 s 4.50% 6 d 4.47% 7 ee 4.46% 8 ie 3.97% 9 l 3.84% 10 a 3.45% 11 th 3.18% 12 z 2.97% 13 k 2.95% 14 r 2.84% 15 m 2.76% 16 e 2.48% 17 w 2.41% 18 er 2.34% 19 v 2.05% 20 ue 2.05% 21 p 1.98% 22 g 1.81% 23 b 1.78% 24 f 1.73% 25 h 1.68% 26 aa 1.55% 27 ae 1.51% 28 oe 1.19% 29 y 1.13% 30 oo 1.04% 31 sh 0.77% 32 or 0.68% 33 thh 0.66% 34 au 0.63% 35 ch 0.59% 36 ou 0.58% 37 air 0.54% 38 j 0.51% 39 oi 0.10% 40 zh 0.05% 100.00% (Note "r" has verb and consonant forms) ===== Read all about truespel at truespel.com. Convert text to truespel USA accent by copy/pasting it at: http://www.foreignword.com/dictionary/truespel/transpel.htm __________________________________________________ Do You Yahoo!? Yahoo! Autos - Get free new car price quotes
This archive was generated by hypermail 2b30 : Fri Jan 17 2003 - 14:46:25 EST