• norsk
    • English
  • norsk 
    • norsk
    • English
  • Logg inn
Vis innførsel 
  •   Hjem
  • Øvrige samlinger
  • Publikasjoner fra CRIStin - NTNU
  • Vis innførsel
  •   Hjem
  • Øvrige samlinger
  • Publikasjoner fra CRIStin - NTNU
  • Vis innførsel
JavaScript is disabled for your browser. Some features of this site may not work without it.

SYNTHETIC DATA FOR DNN-BASED DOA ESTIMATION OF INDOOR SPEECH

Gelderblom, Femke B.; Kvam, Johannes; Liu, Yi; Myrvoll, Tor Andre
Chapter
Accepted version
Thumbnail
Åpne
Gelderblom (198.7Kb)
Permanent lenke
https://hdl.handle.net/11250/2824872
Utgivelsesdato
2021
Metadata
Vis full innførsel
Samlinger
  • Institutt for elektroniske systemer [1865]
  • Publikasjoner fra CRIStin - NTNU [26751]
Originalversjon
10.1109/ICASSP39728.2021.9414415
Sammendrag
This paper investigates the use of different room impulse response (RIR) simulation methods for synthesizing training data for deep neural network-based direction of arrival (DOA) estimation of speech in reverberant rooms. Different sets of synthetic RIRs are obtained using the image source method (ISM) and more advanced methods including diffuse reflections and/or source directivity. Multi-layer perceptron (MLP) deep neural network (DNN) models are trained on generalized cross correlation (GCC) features extracted for each set. Finally, models are tested on features obtained from measured RIRs. This study shows the importance of training with RIRs from directive sources, as resultant DOA models achieved up to 51% error reduction compared to the steered response power with phase transform (SRP-PHAT) baseline (significant with p<<.01), while models trained with RIRs from omnidirectional sources did worse than the baseline. The performance difference was specifically present when estimating the azimuth of speakers not facing the array directly.
Utgiver
Institute of Electrical and Electronics Engineers (IEEE)

Kontakt oss | Gi tilbakemelding

Personvernerklæring
DSpace software copyright © 2002-2019  DuraSpace

Levert av  Unit
 

 

Bla i

Hele arkivetDelarkiv og samlingerUtgivelsesdatoForfattereTitlerEmneordDokumenttyperTidsskrifterDenne samlingenUtgivelsesdatoForfattereTitlerEmneordDokumenttyperTidsskrifter

Min side

Logg inn

Statistikk

Besøksstatistikk

Kontakt oss | Gi tilbakemelding

Personvernerklæring
DSpace software copyright © 2002-2019  DuraSpace

Levert av  Unit