mlsanityData to 7 Oct 2026

The Hitachi-JHU DIHARD III System: Competitive End-to-End Neural Diarization and X-Vector Clustering Systems Combined by DOVER-Lap

arXiv 2102.01363 ↗First posted 2 Feb 2021eess.AS cs.CL cs.SDcited by 8

Not on the map. Why?

Abstract (arXiv)

This paper provides a detailed description of the Hitachi-JHU system that was submitted to the Third DIHARD Speech Diarization Challenge. The system outputs the ensemble results of the five subsystems: two x-vector-based subsystems, two end-to-end neural diarization-based subsystems, and one hybrid subsystem. We refine each system and all five subsystems become competitive and complementary. After the DOVER-Lap based system combination, it achieved diarization error rates of 11.58 % and 14.09 % in Track 1 full and core, and 16.94 % and 20.01 % in Track 2 full and core, respectively. With their results, we won second place in all the tasks of the challenge.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 19 of its 28 references that are in the map. First posted on arXiv 2 Feb 2021.

Builds on

19 of 28 references.

References · 19

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,838
  5. 2019SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognitioncited by 1,355
  6. 2015MUSAN: A Music, Speech, and Noise Corpuscited by 400
  7. 2019pyannote.audio: neural building blocks for speaker diarizationcited by 175
  8. 2018Speaker Recognition from Raw Waveform with SincNetcited by 166
  9. 2019End-to-End Neural Speaker Diarization with Self-attentioncited by 74
  10. 2020End-to-End Speaker Diarization for an Unknown Number of Speakers with Encoder-Decoder Based Attractorscited by 70
  11. 2020The Third DIHARD Diarization Challengecited by 44
  12. 2020DOVER-Lap: A Method for Combining Overlap-aware Diarization Outputscited by 18
7 more
  1. 2020Microsoft Speaker Diarization System for the VoxCeleb Speaker Recognition Challenge 2020cited by 17
  2. 2020Neural Speaker Diarization with Speaker-Wise Chain Rulecited by 14
  3. 2020End-to-End Speaker Diarization as Post-Processingcited by 13
  4. 2019End-to-end Domain-Adversarial Voice Activity Detectioncited by 11
  5. 2020BUT System for the Second DIHARD Speech Diarization Challengecited by 11
  6. 2019DOVER: A Method for Combining Diarization Outputscited by 6
  7. 2019BUT System Description for DIHARD Speech Diarization Challenge 2019cited by 5

Built on by

Cited by year

Citing papers per year: 2021: 2; 2022: 2; 2023: 1; 2024: 1; 2025: 2.

8 papers.

  1. 2021Towards Neural Diarization for Unlimited Numbers of Speakers Using Global and Local Attractorscited by 14
  2. 2022Online Neural Diarization of Unlimited Numbers of Speakers Using Global and Local Attractorscited by 7
  3. 2021Adapting Speaker Embeddings for Speaker Diarisationcited by 4
  4. 2025Unifying Diarization, Separation, and ASR with Multi-Speaker Encodercited by 4
  5. 2022Royalflush Speaker Diarization System for ICASSP 2022 Multi-channel Multi-party Meeting Transcription Challengecited by 2
  6. 2023End-to-End Integration of Speech Separation and Voice Activity Detection for Low-Latency Diarization of Telephone Conversationscited by 2
  7. 2024From Modular to End-to-End Speaker Diarizationcited by 0
  8. 2025Speaker Diarization with Overlapping Community Detection Using Graph Attention Networks and Label Propagation Algorithmcited by 0