Spectral Features derived from Single Frequency Filter for Multispeaker Localization

Sushmita Thakallapalli, Sudarsana Kadiri, Suryakanth Gangashetty

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

2 Citations (Scopus)
93 Downloads (Pure)

Abstract

In this paper, we present a multispeaker localization method using the time delay estimates obtained from the spectral features derived from the single frequency filter (SFF) representation. The mixture signals are transformed into
SFF domain from which the temporal envelopes are extracted at each frequency. Subsequently, the spectral features such as mean and variance of temporal envelopes across frequencies are correlated for extracting the time delay estimates. Since these features emphasize the high SNR regions of the mixtures,
correlation of the corresponding features across the channels leads to robust delay estimates in real acoustic environments. We study the efficacy of the developed approach by comparing its performance with the existing correlation based time delay estimation techniques. Both, a standard data set recorded in
real-room acoustic environments and simulated data set are used for evaluations. It is observed that the localization performance of the proposed algorithm closely matches the performance of a state-of-the-art correlation approach and outperforms other approaches.
Original languageEnglish
Title of host publication26th National Conference on Communications, NCC 2020
PublisherIEEE
ISBN (Electronic)978-1-7281-5120-5
ISBN (Print)978-1-7281-5120-5
DOIs
Publication statusPublished - Feb 2020
MoE publication typeA4 Article in a conference publication
EventNational Conference on Communications - Kharagpur, India
Duration: 21 Feb 202023 Feb 2020
Conference number: 26

Conference

ConferenceNational Conference on Communications
Abbreviated titleNCC
Country/TerritoryIndia
CityKharagpur
Period21/02/202023/02/2020

Fingerprint

Dive into the research topics of 'Spectral Features derived from Single Frequency Filter for Multispeaker Localization'. Together they form a unique fingerprint.

Cite this