Skip to Main content Skip to Navigation
Conference papers

Comparison of Deep Co-Training and Mean-Teacher approaches for semi-supervised audio tagging

Abstract : Recently, a number of semi-supervised learning (SSL) methods, in the framework of deep learning (DL), were shown to achieve state-of-the-art results on image datasets, while using a (very) limited amount of labeled data. To our knowledge, these approaches adapted and applied to audio data are still sparse, in particular for audio tagging (AT). In this work, we adapted the Deep-Co-Training algorithm (DCT) to perform AT, and compared it to another SSL approach called Mean Teacher (MT), that has been used by the winning participants of the DCASE competitions these last two years. Experiments were performed on three standard audio datasets: Environmental Sound classification (ESC-10), UrbanSound8K, and Google Speech Commands. We show that both DCT and MT achieved performance approaching that of a fully supervised training setting, while using a fraction of the labeled data available, and the remaining data as unlabeled data. In some cases, DCT even reached the best accuracy, for instance, 72.6% using half of the labeled data, compared to 74.4% using all the labeled data. DCT also consistently outperformed MT in almost all configurations. For instance, the most significant relative gains brought by DCT reached 12.2% on ESC-10, compared to 7.6% with MT.
Complete list of metadata
Contributor : Thomas Pellegrini <>
Submitted on : Tuesday, March 16, 2021 - 9:32:01 AM
Last modification on : Saturday, March 20, 2021 - 3:30:41 AM


Files produced by the author(s)


  • HAL Id : hal-03170277, version 1


Léo Cances, Thomas Pellegrini. Comparison of Deep Co-Training and Mean-Teacher approaches for semi-supervised audio tagging. IEEE 46th International Conference on Acoustics, Speech and Signal Processing (ICASSP 2021), IEEE Signal Processing Society’s, Jun 2021, Toronto, Canada. ⟨hal-03170277⟩



Record views


Files downloads