Identi cation of Independent Kinematic Regions of the Face during Speech Production
نویسندگان
چکیده
This paper reports our progress on the empirical modeling of facial biomechanics during speech production. Our model is based on decomposing the facial surface into a nite set of linearly independent kinematic regions, which is used as a basis to represent the total facial motion. The main algorithm, based on the column-pivoted QR factorization, is reviewed and compared to other techniques commonly used in statistical regression. The results show that the former technique is more robust to variations in the speech data, and detects regions with a higher measure of independency.
منابع مشابه
SECURING INTERPRETABILITY OF FUZZY MODELS FOR MODELING NONLINEAR MIMO SYSTEMS USING A HYBRID OF EVOLUTIONARY ALGORITHMS
In this study, a Multi-Objective Genetic Algorithm (MOGA) is utilized to extract interpretable and compact fuzzy rule bases for modeling nonlinear Multi-input Multi-output (MIMO) systems. In the process of non- linear system identi cation, structure selection, parameter estimation, model performance and model validation are important objectives. Furthermore, se- curing low-level and high-level ...
متن کاملSelective use of the speech spectrum and a VQGMM method for speaker identification
This paper describes two separate sets of speaker identi cation experiments. In the rst set of experiments, the speech spectrum is selectively used for speaker identi cation. The results show that the higher portion of the speech spectrum contains more reliable idiosyncratic information on speakers than does the lower portion of equal bandwidth. In the second set of experiments, a vector-quanti...
متن کاملInferring linguistic structure in spoken language
We demonstrate the applications of Markov Chains and HMMs to modeling of the underlying structure in spontaneous spoken language. Experiments with supervised training cover the detection of the current dialog state and identi cation of the speech act as used by the speech translation component in our JANUS Speech-to-Speech Translation System. HMM training with hidden states is used to uncover o...
متن کاملSpeech compression with preservation of speaker identity
Although much e ort has been directed recently towards speech compression at rates below 4 kb/s, the primary metric for comparison has, understandably, been the amount of spectral distortion in the decompressed speech. However, an aspect which is becoming important in some applications is the ability to identify the original speaker from the coded speech algorithmically. We investigate here the...
متن کاملTheoretical Error Prediction for a Language Identi cation System using Optimal Phoneme Clustering
using Optimal Phoneme Clustering Kay M. Berkling, Etienne Barnard (berkling,barnard)@cse.ogi.edu Center for Spoken Language Understanding, Oregon Graduate Institute of Science and Technology Abstract A neural network based language identi cation system is described, which uses language independent phoneme clusters as speech units to recognize the language spoken by native speakers over the tele...
متن کامل