Attention-based Encoder-Decoder End-to-End Neural Diarization with Embedding Enhancer

TL;DR

This paper introduces an attention-based encoder-decoder neural diarization system with an enhancer module, improving generalization to unseen speakers and achieving state-of-the-art results on multiple benchmarks.

Contribution

The paper proposes a novel AED-EEND model with an enhancer and iterative decoding, addressing speaker permutation and unseen speaker challenges in diarization.

Findings

Achieved state-of-the-art DER on CALLHOME, DIHARD II, and AMI datasets.

Demonstrated effectiveness of the enhancer module and iterative decoding approach.

Showed that training with more realistic simulated data improves model consistency.

Abstract

Deep neural network-based systems have significantly improved the performance of speaker diarization tasks. However, end-to-end neural diarization (EEND) systems often struggle to generalize to scenarios with an unseen number of speakers, while target speaker voice activity detection (TS-VAD) systems tend to be overly complex. In this paper, we propose a simple attention-based encoder-decoder network for end-to-end neural diarization (AED-EEND). In our training process, we introduce a teacher-forcing strategy to address the speaker permutation problem, leading to faster model convergence. For evaluation, we propose an iterative decoding method that outputs diarization results for each speaker sequentially. Additionally, we propose an Enhancer module to enhance the frame-level speaker embeddings, enabling the model to handle scenarios with an unseen number of speakers. We also explore replacing the transformer encoder with a Conformer architecture, which better models local information. Furthermore, we discovered that commonly used simulation datasets for speaker diarization have a much higher overlap ratio compared to real data. We found that using simulated training data that is more consistent with real data can achieve an improvement in consistency. Extensive experimental validation demonstrates the effectiveness of our proposed methodologies. Our best system achieved a new state-of-the-art diarization error rate (DER) performance on all the CALLHOME (10.08%), DIHARD II (24.64%), and AMI (13.00%) evaluation benchmarks, when no oracle voice activity detection (VAD) is used. Beyond speaker diarization, our AED-EEND system also shows remarkable competitiveness as a speech type detection model.