Here is the story . . .
For over 15 years, two of the basic techniques of Haptic Pronunciation Teaching, have been the Syllable Butterfly and the Tai Chi Finger Flow Fluency. (Check out the links to those two from version 2.0.) The Butterfly focused on syllables and rhythm; the Tai Chi, on fluency. In a recent class where learners were trained in the Tai Chi, there was some additional time left at the end where we work on problematic words have noted during the week. The first student had several multi-syllable words, including 'custodial" and "maintenance." There was almost no contrast between stressed and unstressed syllables--and no obvious rhythmic structure. The Tai Chi technique made little difference, other than speeding up his speech a bit.
For the first time ever in working with vocabulary, I had him do the Butterfly first, which exaggerates both stress contrast and rhythm/length of each syllable, and then the Tai Chi. The student and the class were almost blown away by the improvement. He sounded almost . . . Canadian! Actually, his pronunciation of the segmentals, per se, did not change much if at all, but the relative amplitude and length of the syllables certainly did! The same happened on another dozen words or so after that. In subsequent weeks, the final segment of each class has become basically problematic vocabulary done to the tune of the Butterfly and Tai Chi. Have already included the "Fly-Chi" in Lesson 2 of the KINETIK method!
It is well established in the field that if a learner has relatively good rhythm, including contrast between stressed and unstressed vowels, that problems with segmentals (vowels and consonants) should be less . . . problematic. Actually, and more importantly, with unstressed syllables somewhat backgrounded, vowels tend to be "reduced" and consonant near misses are not only not as noticeable but elide (blend with) adjacent sounds. The process is very evident developmentally in child language learning, as well as the efficacy of embodied techniques such as handclapping in facilitating enhancement of segmentals, e.g., Baills and Prieto, (2023.)
There are "billions and billions" of videos on the web demonstrating the phenomenon and the effect. What those demonstrations do not do, however, is show how to help learners work with the process consistently so that the changes "stick." In preparing for a recent class, I seem to have "discovered" a way . . . The problem with either just using hand clapping with sentences or longer texts, such as songs, or focusing on just one word at a time is that the learner generally not able to take the process out of class effectively, consistently, and remember what was temporarily "embodied" in the process.
Students are using the Fly-Chi now in homework. (I have three weeks of good data so far!) And at least initially, their ability to both change and remember targeted words appears greatly enhanced. Just a fascinating "theory" at this point, Help me test it--and report back! 
