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

Abstract

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.

References in corpus (4)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Conformer: Convolution-augmented Transformer for Speech Recognition
• Neural Speaker Diarization with Speaker-Wise Chain Rule
• BUT System Description for DIHARD Speech Diarization Challenge 2019

Cited by in corpus (5)

• Community Detection Graph Convolutional Network for Overlap-Aware Speaker Diarization
• End-to-End Integration of Speech Separation and Voice Activity Detection for Low-Latency Diarization of Telephone Conversations
• Online Streaming End-to-End Neural Diarization Handling Overlapping Speech and Flexible Numbers of Speakers
• Towards Neural Diarization for Unlimited Numbers of Speakers Using Global and Local Attractors
• Semi-Supervised Training with Pseudo-Labeling for End-to-End Neural Diarization