PWorkshop Archives: Autumn Term 2000

07 Nov 2000

Dr. Catherine Mayo

This talk will be a very informal overview of a project currently being carried out in the department. The general goal of the project is to expand our knowledge of how acoustic cue use, or "weighting" develops in normal listeners, and to evaluate two competing hypotheses as to the reasons for developmental changes in cue weighting strategies.

14 Nov 2000

Dr. Tae-Yeoub Jang

The main goal of the study is to improve performance of Korean automatic speech recognition by exploiting the fundamental frequency (F0) of vowels, which is affected by identity of the preceding consonant. The hypothesis is that if the vowel F0 is given, the identification of the consonant can be more accurate. The effect, which I will call the `segmental F0 effect', has been confirmed by many phonetic studies across various languages. Most frequently, the F0 value of a vowel has been suggested to be a cue to the voiced/voiceless distinction of the preceding consonant. In Korean, segmental F0 can be useful for differentiating three manners (lax, tense, and aspirated) of stop and affricate articulation. Earlier phonetic studies have found that F0 of a vowel onset becomes higher after strong stops (eg., tense and aspirated sounds) and lower after lax stops. It is also suggested that this effect is more salient in Korean than European languages like English and French.

If the segmental F0 effect is going to be helpful for speech recognition, it has to be detectable outside the carefully controlled data used for phonetic studies. I show that automatic measurements over a large amount of data can also capture the effect. Other related issues regarding segmental perturbation which have not been dealt with in earlier studies are also investigated. Integration of the segmental F0 effect with speech recognition is achieved using demisyllables as basic recognition units. As some demisyllables are composed of both an onset consonant and the front part of the nucleus, it is relatively easy for them to carry characteristics of the consonant-vowel relation, such as segmental F0, on their own. Besides, I find that an HMM demisyllable based recogniser performs better than a baseline HMM recogniser with phone-like units even before F0 is included. Thus, using demisyllables in Korean speech recognition has an independent motivation. I show that inclusion of F0 in the demisyllable recogniser gives further improvement in results.

21 Nov 2000

Korin Richmond, Joe Frankel, Olga Goubanova

Recommended reading

28 Nov 2000

Dr. Phil Hoole (Ludwig-Maximilians University, Munich)

12 Dec 2000

Dr. Nigel Fabb (University of Strathclyde)

This paper presents a new theory of metrical structure in verse, developed from the account of stress and accent in Idsardi (1992). Metrical elements (syllables and morae) are grouped into feet by the insertion of right or left parentheses. A foot is constituted by the metrical elements to the right of a left parenthesis or to the left of a right parenthesis. A foot is thus defined by a single parenthesis, not by a matched pair of parentheses, as in other metrical theories. Parentheses are inserted into the verse line by special rules belonging to a small, narrowly restricted set. The most important rules insert parentheses iteratively from one end of the line, skipping over two or three elements at a time. Different footings - ie., different meters - result from different choices of parenthesis insertion rules, supplemented by a small number of well-formedness conditions. The central insight of the theory is that meters are based on rudimentary counting algorithms, rather than on template matching. The theory is illustrated and defended by showing that it accounts correctly for metrical verse of all kinds, exemplified by English iambic pentameter, French alexandrin, classical hexameter, as well as Welsh and Irish syllable counting verse.