The 2023 Multilingual Speech Universal Performance Benchmark (ML-SUPERB) Challenge expands on the acclaimed SUPERB framework. It emphasizes self-supervised models in multilingual speech recognition and language identification. The challenge consists of a research track, a Challenge Track for model submissions, and a New Language Track where researchers can contribute and evaluate their low-resource language data. The findings suggest that merely scaling models is not the definitive solution for multilingual speech tasks, and different speech/voice types present significant challenges in multilingual speech processing.
Publication date: 10 Oct 2023
Project Page: Not provided
Paper: https://arxiv.org/pdf/2310.05513
