Analysis of ABC Frontend Audio Systems for the NIST-SRE24

Analysis of ABC Frontend Audio Systems for the NIST-SRE24

Stať ve sborníku v databázi WoS či Scopus

We present a comprehensive analysis of the embedding extractors (frontends) developed by the ABC team for the audio track of NIST SRE 2024. We follow the two scenarios imposed by NIST: using only a provided set of telephone recordings for training (fixed) or adding publicly available data (open condition). Under these constraints, we develop the best possible speaker embedding extractors for the pre-dominant conversational telephone speech (CTS) domain. We explored architectures based on ResNet with different pooling mechanisms, recently introduced ReDimNet architecture, as well as a system based on the XLS-R model, which represents the family of large pre-trained self-supervised models. In open condition, we train on VoxBlink2 dataset, containing 110 thousand speakers across multiple languages. We observed a good performance and robustness of VoxBlink-trained models, and our experiments show practical recipes for developing state-of-the-art frontends for speaker recognition.

We present a comprehensive analysis of the embedding extractors (frontends) developed by the ABC team for the audio track of NIST SRE 2024. We follow the two scenarios imposed by NIST: using only a provided set of telephone recordings for training (fixed) or adding publicly available data (open condition). Under these constraints, we develop the best possible speaker embedding extractors for the pre-dominant conversational telephone speech (CTS) domain. We explored architectures based on ResNet with different pooling mechanisms, recently introduced ReDimNet architecture, as well as a system based on the XLS-R model, which represents the family of large pre-trained self-supervised models. In open condition, we train on VoxBlink2 dataset, containing 110 thousand speakers across multiple languages. We observed a good performance and robustness of VoxBlink-trained models, and our experiments show practical recipes for developing state-of-the-art frontends for speaker recognition.

embedding extractors | NIST-SRE | speaker recognition | VoxBlink

embedding extractors | NIST-SRE | speaker recognition | VoxBlink

BARAHONA, S.; SILNOVA, A.; MOŠNER, L.; PENG, J.; PLCHOT, O.; ROHDIN, J.; ZHANG, L.; HAN, J.; PALKA, P.; LANDINI, F.; BURGET, L.; STAFYLAKIS, T.; CUMANI, S.; BOBOŠ, D.; HLAVAČEK, M.; KODOVSKY, M.; PAVLIČEK, T.

2026

17.08.2025

International Speech Communication Association

Rotterdam

Proceedings of the Annual Conference of the International Speech Communication Association Interspeech

Interspeech

Nizozemsko

5763

5767

5

https://www.isca-archive.org/interspeech_2025/barahona25_interspeech.pdf

barahona_2025_interspeech