TalTech-IRIT-LIS Speaker and Language Diarization Systems for DISPLACE 2024

TL;DR

This paper presents TalTech-IRIT-LIS's systems for speaker and language diarization in DISPLACE 2024, achieving top rankings with novel ensemble and embedding techniques.

Contribution

The paper introduces ensemble speaker diarization combining PixIT with separation outputs and fine-tunes Wav2Vec2-BERT for language diarization, advancing state-of-the-art results.

Findings

Achieved 27.1% diarization error rate in speaker diarization.

Achieved 27.6% language diarization error rate.

Ranked first in both challenge tracks.

Abstract

This paper describes the submissions of team TalTech-IRIT-LIS to the DISPLACE 2024 challenge. Our team participated in the speaker diarization and language diarization tracks of the challenge. In the speaker diarization track, our best submission was an ensemble of systems based on the pyannote.audio speaker diarization pipeline utilizing powerset training and our recently proposed PixIT method that performs joint diarization and speech separation. We improve upon PixIT by using the separation outputs for speaker embedding extraction. Our ensemble achieved a diarization error rate of 27.1% on the evaluation dataset. In the language diarization track, we fine-tuned a pre-trained Wav2Vec2-BERT language embedding model on in-domain data, and clustered short segments using AHC and VBx, based on similarity scores from LDA/PLDA. This led to a language diarization error rate of 27.6% on the evaluation data. Both results were ranked first in their respective challenge tracks.