Information retrieval from a phoneme time series database

TL;DR

This paper introduces a novel method for retrieving phoneme time series from a database by transforming the problem into sequence retrieval using phase-space partitioning, and evaluates its robustness against noise.

Contribution

It proposes a new approach that maps time series retrieval to sequence retrieval via k-means clustering in phase-space, addressing limitations of classical spectral methods.

Findings

The method effectively retrieves phoneme sequences even with noise.

Classical power-spectral techniques are less effective for this task.

The approach handles both whole and subsequence matching.

Abstract

Developing fast and efficient algorithms for retrieval of objects to a given user query is an area of active research. The present study investigates retrieval of time series objects from a phoneme database to a given user pattern or query. The proposed method maps the one-dimensional time series retrieval into a sequence retrieval problem by partitioning the multi-dimensional phase-space using k-means clustering. The problem of whole sequence as well as subsequence matching is considered. Robustness of the proposed technique is investigated on phoneme time series corrupted with additive white Gaussian noise. The shortcoming of classical power-spectral techniques for time series retrieval is also discussed.