Speech Intelligibility Improvement in Noisy Environments for Near-End Listening Enhancement
Journal Title: Journal of Information Systems and Telecommunication - Year 2016, Vol 4, Issue 1
Abstract
A new speech intelligibility improvement method for near-end listening enhancement in noisy environments is proposed. This method improves speech intelligibility by optimizing energy correlation of one-third octave bands of clean speech and enhanced noisy speech without power increasing. The energy correlation is determined as a cost function based on frequency band gains of the clean speech. Interior-point algorithm which is an iterative procedure for the nonlinear optimization is used to determine the optimal points of the cost function because of nonlinearity and complexity of the energy correlation function. Two objective intelligibility measures, speech intelligibility index and short-time objective intelligibility measure, are employed to evaluate the noisy enhanced speech intelligibility. Furthermore, the speech intelligibility scores are compared with unprocessed speech and a baseline method under various noisy conditions. The results show large intelligibility improvements with the proposed method over the unprocessed noisy speech.
Authors and Affiliations
Peyman Goli, Mohammad Reza Karami-Mollaei
A Fast and Accurate Sound Source Localization Method using Optimal Combination of SRP and TDOA Methodologies
This paper presents an automatic sound source localization approach based on combination of the basic time delay estimation sub method namely, Time Difference of Arrival (TDOA), and Steered Response Power (SRP) methods....
A Low-Jitter 20-110MHz DLL Based on a Simple PD and Common-Mode Voltage Level Corrected Differential Delay Elements
In this paper, a 16-phases 20MHz to 110MHz low jitter delay locked loop, DLL, is proposed in a 0.35µm CMOS process. A sensitive open loop phase detector, PD, is introduced based on a novel idea to simply detect small pha...
Application of Curve Fitting in Hyperspectral Data Classification and Compression
Regarding to the high between-band correlation and large volumes of hyperspectral data, feature reduction (either feature selection or extraction) is an important part of classification process for this data type. A vari...
Unsupervised Segmentation of Retinal Blood Vessels Using the Human Visual System Line Detection Model
Retinal image assessment has been employed by the medical community for diagnosing vascular and non-vascular pathology. Computer based analysis of blood vessels in retinal images will help ophthalmologists monitor larger...
A New Recursive Algorithm for Universal Coding of Integers
In this paper, we aim to encode the set of all positive integers so that the codewords not only be uniquely decodable but also be an instantaneous set of binary sequences. Elias introduces three recursive algorithms for...