Phase-space representation of speech
نویسنده
چکیده
Speech production is essentially a nonlinear dynamic process. Motivated by ideas in dynamic system research, this paper seeks to recast the speech representation problem (front-end) as an attempt to reconstruct the phase space of the production process, or articulatory configurations. We point out that the use of the delta and double delta features, common in current ASR (Automatic Speech Recognition) systems, corresponds to timedelayed embedding, a technique in nonlinear time series analysis for phase space reconstruction. The traditional delta and double features also impose a suboptimal linear transform in the reconstructed space. We show that a significant improvement in recognition accuracy can be achieved by choosing the transform in a data-driven fashion.
منابع مشابه
Mental Timeline in Persian Speakers’ Co-speech Gestures based on Lakoff and Johnson’s Conceptual Metaphor Theory
One of the introduced conceptual metaphors is the metaphor of "time as space". Time as an abstract concept is conceptualized by a concrete concept like space. This conceptualization of time is also reflected in co-speech gestures. In this research, we try to find out what dimension and direction the mental timeline has in co-speech gestures and under the influence of which one of the metaphoric...
متن کاملPhase-space Representation of Speech — Revisiting the Delta and Double Delta Features
Speech production is essentially a nonlinear dynamic process. Motivated by ideas in dynamic system research, this paper seeks to recast the speech representation problem (frontend) as an attempt to reconstruct the phase space of the production process, or articulatory configurations. In particular, we point out that the use of the delta and double delta features, common in current ASR (Automati...
متن کاملFast Reconstruction of SAR Images with Phase Error Using Sparse Representation
In the past years, a number of algorithms have been introduced for synthesis aperture radar (SAR) imaging. However, they all suffer from the same problem: The data size to process is considerably large. In recent years, compressive sensing and sparse representation of the signal in SAR has gained a significant research interest. This method offers the advantage of reducing the sampling rate, bu...
متن کاملVoice-based Age and Gender Recognition using Training Generative Sparse Model
Abstract: Gender recognition and age detection are important problems in telephone speech processing to investigate the identity of an individual using voice characteristics. In this paper a new gender and age recognition system is introduced based on generative incoherent models learned using sparse non-negative matrix factorization and atom correction post-processing method. Similar to genera...
متن کاملdominating subset and representation graph on topological spaces
Let a topological space. An intersection graph on a topological space , which denoted by , is an undirected graph which whose vertices are open subsets of and two vertices are adjacent if the intersection of them are nonempty. In this paper, the relation between topological properties of and graph properties of are investigated. Also some classifications and representations for the graph ...
متن کامل