Vers la compréhension automatique de la parole bout-en-bout à moindre effort
This work addresses the efficiency problem for researchers and practitioners using French speech models, but it is incremental as it builds on existing methods.
The paper tackled the high computational and energy costs of self-supervised models for spoken language understanding in French by comparing learning strategies to reduce these costs, achieving state-of-the-art performance on the MEDIA corpus.
Recent advances in spoken language understanding benefited from Self-Supervised models trained on large speech corpora. For French, the LeBenchmark project has made such models available and has led to impressive progress on several tasks including spoken language understanding. These advances have a non-negligible cost in terms of computation time and energy consumption. In this paper, we compare several learning strategies aiming at reducing such cost while keeping competitive performances. The experiments are performed on the MEDIA corpus, and show that it is possible to reduce the learning cost while maintaining state-of-the-art performances.