# Neural Diarization with Non-autoregressive Intermediate Attractors

arXiv: https://arxiv.org/abs/2303.06806 · HTML: https://mlsanity.com/paper/2303.06806/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 33 references that are in the map.
- First posted on arXiv 13 Mar 2023.

## Builds on (most cited first, 21 of 21)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [MUSAN: A Music, Speech, and Noise Corpus](https://mlsanity.com/paper/1510.08484/) (2015)
- [Generalized End-to-End Loss for Speaker Verification](https://mlsanity.com/paper/1710.10467/) (2017)
- [Deep clustering: Discriminative embeddings for segmentation and separation](https://mlsanity.com/paper/1508.04306/) (2015)
- [Permutation Invariant Training of Deep Models for Speaker-Independent Multi-talker Speech Separation](https://mlsanity.com/paper/1607.00325/) (2016)
- [A Review of Speaker Diarization: Recent Advances with Deep Learning](https://mlsanity.com/paper/2101.09624/) (2021)
- [Intermediate Loss Regularization for CTC-based Speech Recognition](https://mlsanity.com/paper/2102.03216/) (2021)
- [End-to-End Neural Speaker Diarization with Permutation-Free Objectives](https://mlsanity.com/paper/1909.05952/) (2019)
- [Spot the conversation: speaker diarisation in the wild](https://mlsanity.com/paper/2007.01216/) (2020)
- [End-to-End Neural Speaker Diarization with Self-attention](https://mlsanity.com/paper/1909.06247/) (2019)
- [Speaker Diarization with LSTM](https://mlsanity.com/paper/1710.10468/) (2017)
- [End-to-End Speaker Diarization for an Unknown Number of Speakers with Encoder-Decoder Based Attractors](https://mlsanity.com/paper/2005.09921/) (2020)
- [Relaxing the Conditional Independence Assumption of CTC-based ASR by Conditioning on Intermediate Predictions](https://mlsanity.com/paper/2104.02724/) (2021)
- [The Third DIHARD Diarization Challenge](https://mlsanity.com/paper/2012.01477/) (2020)
- [The Second DIHARD Diarization Challenge: Dataset, task, and baselines](https://mlsanity.com/paper/1906.07839/) (2019)
- [A Comparative Study on Non-Autoregressive Modelings for Speech-to-Text Generation](https://mlsanity.com/paper/2110.05249/) (2021)
- [Encoder-Decoder Based Attractors for End-to-End Neural Diarization](https://mlsanity.com/paper/2106.10654/) (2021)
- [From Simulated Mixtures to Simulated Conversations as Training Data for End-to-End Neural Diarization](https://mlsanity.com/paper/2204.00890/) (2022)
- [End-to-end Neural Diarization: From Transformer to Conformer](https://mlsanity.com/paper/2106.07167/) (2021)
- [End-to-End Diarization for Variable Number of Speakers with Local-Global Networks and Discriminative Speaker Embeddings](https://mlsanity.com/paper/2105.02096/) (2021)
- [Auxiliary Loss of Transformer with Residual Connection for End-to-End Speaker Diarization](https://mlsanity.com/paper/2110.07116/) (2021)

## Built on by (most cited first, 3 of 3)

- [DiaCorrect: Error Correction Back-end For Speaker Diarization](https://mlsanity.com/paper/2309.08377/) (2023)
- [Speakers Unembedded: Embedding-free Approach to Long-form Neural Diarization](https://mlsanity.com/paper/2406.18679/) (2024)
- [Improving Neural Diarization through Speaker Attribute Attractors and Local Dependency Modeling](https://mlsanity.com/paper/2506.05593/) (2025)

## Abstract (arXiv)

End-to-end neural diarization (EEND) with encoder-decoder-based attractors (EDA) is a promising method to handle the whole speaker diarization problem simultaneously with a single neural network. While the EEND model can produce all frame-level speaker labels simultaneously, it disregards output label dependency. In this work, we propose a novel EEND model that introduces the label dependency between frames. The proposed method generates non-autoregressive intermediate attractors to produce speaker labels at the lower layers and conditions the subsequent layers with these labels. While the proposed model works in a non-autoregressive manner, the speaker labels are refined by referring to the whole sequence of intermediate labels. The experiments with the two-speaker CALLHOME dataset show that the intermediate labels with the proposed non-autoregressive intermediate attractors boost the diarization performance. The proposed method with the deeper network benefits more from the intermediate labels, resulting in better performance and training throughput than EEND-EDA.