Exploring wav2vec 2.0 on speaker verification and language identification
This work provides a strong baseline for applying self-supervised speech models to speaker and language recognition, which is significant for researchers working on these specific speech tasks.
This paper explores the application of wav2vec 2.0, a self-supervised speech representation learning framework, to speaker verification and language identification tasks. It achieves a new state-of-the-art Equal Error Rate (EER) of 3.61% on the VoxCeleb1 dataset for speaker verification and EERs of 12.02% (1s) and 3.47% (full-length) on the AP17-OLR dataset for language identification.
Wav2vec 2.0 is a recently proposed self-supervised framework for speech representation learning. It follows a two-stage training process of pre-training and fine-tuning, and performs well in speech recognition tasks especially ultra-low resource cases. In this work, we attempt to extend self-supervised framework to speaker verification and language identification. First, we use some preliminary experiments to indicate that wav2vec 2.0 can capture the information about the speaker and language. Then we demonstrate the effectiveness of wav2vec 2.0 on the two tasks respectively. For speaker verification, we obtain a new state-of-the-art result, Equal Error Rate (EER) of 3.61% on the VoxCeleb1 dataset. For language identification, we obtain an EER of 12.02% on 1 second condition and an EER of 3.47% on full-length condition of the AP17-OLR dataset. Finally, we utilize one model to achieve the unified modeling by the multi-task learning for the two tasks.