We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

eess.AS

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Electrical Engineering and Systems Science > Audio and Speech Processing

Title: The Third DIHARD Diarization Challenge

Abstract: DIHARD III was the third in a series of speaker diarization challenges intended to improve the robustness of diarization systems to variability in recording equipment, noise conditions, and conversational domain. Speaker diarization was evaluated under two speech activity conditions (diarization from a reference speech activity vs. diarization from scratch) and 11 diverse domains. The domains span a range of recording conditions and interaction types, including read audio-books, meeting speech, clinical interviews, web videos, and, for the first time, conversational telephone speech. A total of 30 organizations (forming 21teams) from industry and academia submitted 499 valid system outputs. The evaluation results indicate that speaker diarization has improved markedly since DIHARD I, particularly for two-party interactions, but that for many domains (e.g., web video) the problem remains far from solved.
Comments: arXiv admin note: text overlap with arXiv:1906.07839
Subjects: Audio and Speech Processing (eess.AS); Sound (cs.SD)
Cite as: arXiv:2012.01477 [eess.AS]
  (or arXiv:2012.01477v3 [eess.AS] for this version)

Submission history

From: Neville Ryant [view email]
[v1] Wed, 2 Dec 2020 19:33:44 GMT (30kb,D)
[v2] Wed, 27 Jan 2021 23:42:39 GMT (30kb,D)
[v3] Mon, 5 Apr 2021 17:23:19 GMT (415kb,D)

Link back to: arXiv, form interface, contact.