Robust acoustic domain identification with its application to speaker diarization - Inria - Institut national de recherche en sciences et technologies du numérique
Journal Articles International Journal of Speech Technology Year : 2022

Robust acoustic domain identification with its application to speaker diarization

Abstract

With the rise in multimedia content over the years, more variety is observed in the recording environments of audio. An audio processing system might benefit when it has a module to identify the acoustic domain at its front-end. In this paper, we demonstrate the idea of acoustic domain identification (ADI) for speaker diarization. For this, we first present a detailed study of the various domains of the third DIHARD challenge highlighting the factors that differentiated them from each other. Our main contribution is to develop a simple and efficient solution for ADI. In the present work, we explore speaker embeddings for this task. Next, we integrate the ADI module with the speaker diarization framework of the DIHARD III challenge. The performance substantially improved over that of the baseline when the thresholds for agglomerative hierarchical clustering were optimized according to the respective domains. We achieved a relative improvement of more than 5% and 8% in DER for core and full conditions, respectively, on Track 1 of the DIHARD III evaluation set.
Fichier principal
Vignette du fichier
IJST_DIHARD3_Submitted.pdf (403.43 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03719697 , version 1 (11-07-2022)

Identifiers

Cite

A Kishore Kumar, Shefali Waldekar, Md Sahidullah, Goutam Saha. Robust acoustic domain identification with its application to speaker diarization. International Journal of Speech Technology, 2022, 25 (December), pp.933-945. ⟨10.1007/s10772-022-09990-9⟩. ⟨hal-03719697⟩
66 View
113 Download

Altmetric

Share

More