mlsanityData to 14 Oct 2026

Improving the Naturalness of Simulated Conversations for End-to-End Neural Diarization

arXiv 2204.11232 ↗First posted 24 Apr 2022eess.AS cs.CL cs.SDcited by 7

Not on the map. Why?

Abstract (arXiv)

This paper investigates a method for simulating natural conversation in the model training of end-to-end neural diarization (EEND). Due to the lack of any annotated real conversational dataset, EEND is usually pretrained on a large-scale simulated conversational dataset first and then adapted to the target real dataset. Simulated datasets play an essential role in the training of EEND, but as yet there has been insufficient investigation into an optimal simulation method. We thus propose a method to simulate natural conversational speech. In contrast to conventional methods, which simply combine the speech of multiple speakers, our method takes turn-taking into account. We define four types of speaker transition and sequentially arrange them to simulate natural conversations. The dataset simulated using our method was found to be statistically similar to the real dataset in terms of the silence and overlap ratios. The experimental results on two-speaker diarization using the CALLHOME and CSJ datasets showed that the simulated dataset contributes to improving the performance of EEND.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 29 of its 48 references that are in the map. First posted on arXiv 24 Apr 2022.

Builds on

29 of 48 references.

References · 29

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,863
  4. 2019SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognitioncited by 1,380
  5. 2015MUSAN: A Music, Speech, and Noise Corpuscited by 425
  6. 2015Deep clustering: Discriminative embeddings for segmentation and separationcited by 212
  7. 2019WHAM!: Extending Speech Separation to Noisy Environmentscited by 156
  8. 2020LibriMix: An Open-Source Dataset for Generalizable Speech Separationcited by 155
  9. 2021A Review of Speaker Diarization: Recent Advances with Deep Learningcited by 124
  10. 2018The fifth 'CHiME' Speech Separation and Recognition Challenge: Dataset, task and baselinescited by 92
  11. 2019End-to-End Neural Speaker Diarization with Permutation-Free Objectivescited by 88
  12. 2020Serialized Output Training for End-to-End Overlapped Speech Recognitioncited by 87
17 more
  1. 2019End-to-End Neural Speaker Diarization with Self-attentioncited by 77
  2. 2020Continuous speech separation: dataset and analysiscited by 74
  3. 2020Bayesian HMM clustering of x-vector sequences (VBx) in speaker diarization: theory, implementation and analysis on standard taskscited by 73
  4. 2020End-to-End Speaker Diarization for an Unknown Number of Speakers with Encoder-Decoder Based Attractorscited by 71
  5. 2019WHAMR!: Noisy and Reverberant Single-Channel Speech Separationcited by 61
  6. 2018Fully Supervised Speaker Diarizationcited by 47
  7. 2020Auto-Tuning Spectral Clustering for Speaker Diarization Using Normalized Maximum Eigengapcited by 36
  8. 2018A Purely End-to-end System for Multi-speaker Speech Recognitioncited by 30
  9. 2019Advances in Online Audio-Visual Meeting Transcriptioncited by 23
  10. 2021Encoder-Decoder Based Attractors for End-to-End Neural Diarizationcited by 20
  11. 2019Discriminative Neural Clustering for Speaker Diarisationcited by 14
  12. 2019Simultaneous Speech Recognition and Speaker Diarization for Monaural Dialogue Recordings with Target-Speaker Acoustic Modelscited by 13
  13. 2021End-to-end Neural Diarization: From Transformer to Conformercited by 12
  14. 2020Tackling real noisy reverberant meetings with all-neural source separation, counting, and diarization systemcited by 10
  15. 2021Hypothesis Stitcher for End-to-End Speaker-attributed ASR on Long-form Multi-talker Recordingscited by 7
  16. 2021End-to-End Diarization for Variable Number of Speakers with Local-Global Networks and Discriminative Speaker Embeddingscited by 7
  17. 2021Semi-Supervised Training with Pseudo-Labeling for End-to-End Neural Diarizationcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 1; 2024: 1; 2025: 1; 2026: 3.

7 papers.

  1. 2022Simulating realistic speech overlaps improves multi-talker ASRcited by 10
  2. 2024Mamba-based Segmentation Model for Speaker Diarizationcited by 4
  3. 2023DiaCorrect: Error Correction Back-end For Speaker Diarizationcited by 3
  4. 2026Speaker-Aware Simulation Improves Conversational Speech Recognitioncited by 2
  5. 2025LibriConvo: Simulating Conversations from Read Literature for ASR and Diarizationcited by 1
  6. 2026Efficient ASR Training with Conversations that Never Happenedcited by 0
  7. 2026On the Role of Conversational Timing in Synthetic Training Data for ASRcited by 0