Title | ||
---|---|---|
A hybrid physical and statistical dynamic articulatory framework incorporating analysis-by-synthesis for improved phone classification |
Abstract | ||
---|---|---|
In this paper, we present a dynamic articulatory model for phone classification. The model integrates real articulatory information de- rived from ElectroMagnetic Articulograph (EMA) data into its inner states. It maps from the articulatory space to the acoustic one using an adapted vocal tract model for each speaker and a physiologically- motivated articulatory synthesis approach. We apply the analysis- by-synthesis paradigm in a statistical fashion. We first present a fast approach for deriving analysis-by-synthesis distortion features. Next, the distortion between the speech synthesized from the artic- ulatory states and the incoming speech signal is used to compute the output observation probabilities of the Hidden Markov Model (HMM) used for classification. Experiments with the novel frame- work show improvements over baseline in phone classification accu- racy. Index Terms—Dynamic articulatory modeling, analysis-by- synthesis, articulatory synthesis for recognition, physical model of the vocal tract, hybrid physical and statistical models for classifica- tion |
Year | DOI | Venue |
---|---|---|
2010 | 10.1109/ICASSP.2010.5495696 | Acoustics Speech and Signal Processing |
Keywords | Field | DocType |
hidden Markov models,speech recognition,speech synthesis,analysis-by-synthesis distortion features,automatic speech recognition systems,electromagnetic articulograph,hidden Markov model,phone classification,physical dynamic articulatory framework,physiologically-motivated articulatory synthesis,statistical dynamic articulatory framework,Dynamic articulatory modeling,analysis-by-synthesis,articulatory synthesis for recognition,hybrid physical,physical model of the vocal tract,statistical models for classification | Speech synthesis,Speech coding,Pattern recognition,Computer science,Articulatory synthesis,Speech recognition,Phone,Artificial intelligence,Hidden Markov model,Distortion,Vocal tract | Conference |
ISSN | ISBN | Citations |
1520-6149 E-ISBN : 978-1-4244-4296-6 | 978-1-4244-4296-6 | 0 |
PageRank | References | Authors |
0.34 | 6 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ziad Al Bawab | 1 | 25 | 2.93 |
Raj, Bhiksha | 2 | 2094 | 204.63 |
Richard M. Stern | 3 | 1663 | 406.79 |