TS-SUPERB: A Target Speech Processing Benchmark for Speech Self-Supervised Learning Models

TS-SUPERB: A Target Speech Processing Benchmark for Speech Self-Supervised Learning Models

Stať ve sborníku v databázi WoS či Scopus

Self-supervised learning (SSL) models have significantly advanced speech processing tasks, and several benchmarks have been pro- posed to validate their effectiveness. However, previous benchmarks have primarily focused on single-speaker scenarios, with less exploration of target-speaker tasks in noisy, multi-talker conditions-a more challenging yet practical case. In this paper, we introduce the Target-Speaker Speech Processing Universal Performance Benchmark (TS-SUPERB), which includes four widely recognized target-speaker processing tasks that require identifying the target speaker and extracting information from the speech mixture. In our benchmark, the speaker embedding extracted from enrollment speech is used as a clue to condition downstream models. The benchmark result reveals the importance of evaluating SSL models in target speaker scenarios, demonstrating that performance cannot be easily inferred from related single-speaker tasks. Moreover, by using a unified SSL-based target speech encoder, consisting of a speaker encoder and an extractor module, we also investigate joint optimization across TS tasks to leverage mutual information and demonstrate its effectiveness.

Self-supervised learning (SSL) models have significantly advanced speech processing tasks, and several benchmarks have been pro- posed to validate their effectiveness. However, previous benchmarks have primarily focused on single-speaker scenarios, with less exploration of target-speaker tasks in noisy, multi-talker conditions-a more challenging yet practical case. In this paper, we introduce the Target-Speaker Speech Processing Universal Performance Benchmark (TS-SUPERB), which includes four widely recognized target-speaker processing tasks that require identifying the target speaker and extracting information from the speech mixture. In our benchmark, the speaker embedding extracted from enrollment speech is used as a clue to condition downstream models. The benchmark result reveals the importance of evaluating SSL models in target speaker scenarios, demonstrating that performance cannot be easily inferred from related single-speaker tasks. Moreover, by using a unified SSL-based target speech encoder, consisting of a speaker encoder and an extractor module, we also investigate joint optimization across TS tasks to leverage mutual information and demonstrate its effectiveness.

Self-supervised learning, target-speaker speech process, speech recognition, speech enhancement, voice activity detection

Self-supervised learning, target-speaker speech process, speech recognition, speech enhancement, voice activity detection

PENG, J.; ASHIHARA, T.; DELCROIX, M.; OCHIAI, T.; PLCHOT, O.; ARAKI, S.; ČERNOCKÝ, J.

06.04.2025

IEEE Signal Processing Society

Hyderabad

979-8-3503-6874-1

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

1

5

5

https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10887574

TS-SUPERB_A_Target_Speech_Processing_Benchmark_for_Speech_Self-Supervised_Learning_Models