How to (Virtually) Train Your Speaker Localizer - INRIA - Institut National de Recherche en Informatique et en Automatique
Conference Papers Year : 2023

How to (Virtually) Train Your Speaker Localizer

Abstract

Learning-based methods have become ubiquitous in speaker localization. Existing systems rely on simulated training sets for the lack of sufficiently large, diverse and annotated real datasets. Most room acoustics simulators used for this purpose rely on the image source method (ISM) because of its computational efficiency. This paper argues that carefully extending the ISM to incorporate more realistic surface, source and microphone responses into training sets can significantly boost the real-world performance of speaker localization systems. It is shown that increasing the training-set realism of a state-of-the-art direction-of-arrival estimator yields consistent improvements across three different real test sets featuring human speakers in a variety of rooms and various microphone arrays. An ablation study further reveals that every added layer of realism contributes positively to these improvements.
Fichier principal
Vignette du fichier
INTERSPEECH_2023-CR.pdf (193.19 Ko) Télécharger le fichier
Origin Files produced by the author(s)
licence

Dates and versions

hal-03855912 , version 1 (21-11-2022)
hal-03855912 , version 2 (30-11-2022)
hal-03855912 , version 3 (25-05-2023)

Licence

Identifiers

  • HAL Id : hal-03855912 , version 3

Cite

Prerak Srivastava, Antoine Deleforge, Archontis Politis, Emmanuel Vincent. How to (Virtually) Train Your Speaker Localizer. INTERSPEECH 2023, Aug 2023, Dublin, Ireland. ⟨hal-03855912v3⟩
203 View
346 Download

Share

More