• Corpus ID: 14390199

Analysis of Voice-Onset Using Active Rays and Hidden Markov Models

@inproceedings{Jesorsky1997AnalysisOV,
  title={Analysis of Voice-Onset Using Active Rays and Hidden Markov Models},
  author={Oliver Jesorsky and Joachim Denzler and T. Wittenberg},
  year={1997}
}
In this paper we describe an approach for an automatic evaluation system for laryngoscopical image sequences. We describe how glottis segmentation in high speed camera sequences can be done by using active rays. We show that active ray contour segmentation provides a good base for useful feature calculation. Afterwards we show how hidden Markov models (HMMs) can serve for a statistical evaluation of the given feature vector sequences. The main advantage of the HMM approach is the fact, that… 

References

SHOWING 1-8 OF 8 REFERENCES

An introduction to hidden Markov models

The purpose of this tutorial paper is to give an introduction to the theory of Markov models, and to illustrate how they have been applied to problems in speech recognition.

Direct evaluation of high-speed recordings of vocal fold vibrations.

A digital high-speed camera system for the endoscopic examination of the larynx allows recording speeds of up to 5,600 frames/s and combined with an image processing program the system is able to render x-t diagrams of vocal cord movement.

Deformation Analysis of the Vibrational Patterns of the Vocal Folds

In this paper, an active contours (snakes) based approach is used to delineate the vocal folds in each frame of the videostroboscopic image sequence and a new elastic registration algorithms are used to register the vocal fold contours between adjacent frames of the video sequence.

Active Rays: A New Approach to Contour Tracking

A new approach to contour extraction and tracking, which is based on the principles of active contour models and overcomes its shortcomings and shows any{time behavior which is important with respect to real{time applications.

Estimation of Vocal Fold Vibrations Using Image Segmentation

A new high speed system for recording, processing, and analysis of vocal fold vibrations has been developed. Results obtained with this system will be discussed from the technical as well as the

Statistische Spracherkennung

Die Erkennungsphase in einem Dialogsystem

Hidden Markov Models for Speech Recognition. Number 7 in Information Technology Series

  • 1990