Search

Purpose: The use of gated or ECG triggered MR is a well-established technique, and developments in coil technology have enabled this approach to be applied to areas other than the heart. However, the image quality of gated (ECG or cine) versus non-gated or real-time has not been extensively evaluated in the mouth. We evaluate two image sequences by developing an automatic image processing technique which compares how well the image represents known anatomy. Methods: Four subjects practised experimental poly-syllabic sentences prior to MR scanning. Using a 1.5 T MR unit, we acquired comparable gated (using an artificial trigger) and non-gated sagittal images during speech. We then used an image processing algorithm to model the image grey along lines that cross the airway. This involved an eight parameter non-linear equation to model of proton densities, edges, and dimensions. Results: Gated and non-gated images show similar spatial resolution, with non-gated images being slightly sharper (10% better resolution, less than 1 pixel). However, the gated sequences generated images of substantially lower inherent noise, and substantially better discrimination between air and tissue. Additionally, the gated sequences demonstrate a very much greater temporal resolution. Conclusion: Overall, image quality is better with gated imaging techniques, especially given their superior temporal resolution. Gated techniques are limited by the repeatability of the motions involved, and we have shown that speech to a metronome can be sufficiently repeatable to allow high-quality gated magnetic resonance imaging images. We suggest that gated sequences may be useful for evaluating other types of repetitive movement involving the joints and limb motions.

We use Stem-ML to build an automatic learning system for Mandarin prosody that allows us to make quantitative measurements for prosodic strengths. Stem-ML is a phenomenological model of the muscle dynamics and planning process that controls the tension of the vocal folds. Because Stem-ML describes the interactions between nearby tones and accents, we were able to use a highly constrained model with only one accent template for each lexical tone category, and a single prosodic strength per word. The model accurately reproduces the intonation of the speaker, capturing 87% of the variance of the speech's fundamental frequency, f0. The result reveals strong alternating metrical patterns in words, and suggests that the speaker uses word strength to mark a hierarchy of sentence, clause, phrase, and word boundaries.

We present a step towards a mathematical description of style in speech and singing. We represent a style of speech or singing as a set of portable prosodic features along with a set of rules to choose where the features are to be applied. Speech and singing can both be conveniently described by separating their style from their content. In both speech and singing, style is implemented in terms of acoustic parameters like pitch and loudness. The features we use correspond to musical embellishments or prosodic accents and phrase curve in speech, and are implemented via Stem-ML (Soft TEMplate Mark-up Language). We present several examples to illustrate the usefulness of the style/content separation, and the similarity between speech and music.