# Transformer Attractors for Robust and Efficient End-to-End Neural Diarization

arXiv: https://arxiv.org/abs/2312.06253 · HTML: https://mlsanity.com/paper/2312.06253/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Dec 2023.

## Built on by (most cited first, 2 of 2)

- [Speakers Unembedded: Embedding-free Approach to Long-form Neural Diarization](https://mlsanity.com/paper/2406.18679/) (2024)
- [Configurable Multilingual ASR with Speech Summary Representations](https://mlsanity.com/paper/2410.04478/) (2024)

## Abstract (arXiv)

End-to-end neural diarization with encoder-decoder based attractors (EEND-EDA) is a method to perform diarization in a single neural network. EDA handles the diarization of a flexible number of speakers by using an LSTM-based encoder-decoder that generates a set of speaker-wise attractors in an autoregressive manner. In this paper, we propose to replace EDA with a transformer-based attractor calculation (TA) module. TA is composed of a Combiner block and a Transformer decoder. The main function of the combiner block is to generate conversational dependent (CD) embeddings by incorporating learned conversational information into a global set of embeddings. These CD embeddings will then serve as the input for the transformer decoder. Results on public datasets show that EEND-TA achieves 2.68% absolute DER improvement over EEND-EDA. EEND-TA inference is 1.28 times faster than that of EEND-EDA.