Abstract
Methods for extracting audio and speech features have been studied since pioneering work on spectrum analysis decades ago. Recent efforts are guided by the ambition to develop general-purpose audio representations. For example, deep neural networks can extract optimal embeddings if they are trained on large audio datasets. This work extends existing methods based on self-supervised learning by bootstrapping, proposes various encoder architectures, and explores the effects of using different pre-training datasets. Lastly, we present a novel training framework to come up with a hybrid audio representation, which combines handcrafted and data-driven learned audio features. All the proposed representations were evaluated within the HEAR NeurIPS 2021 challenge for auditory scene classification and timestamp detection tasks. Our results indicate that the hybrid model with a convolutional transformer as the encoder yields superior performance in most HEAR challenge tasks.
Original language | Undefined/Unknown |
---|---|
Publication date | 24 Jun 2022 |
Publication status | Published - 24 Jun 2022 |
Externally published | Yes |
Event | HEAR: Holistic Evaluation of Audio Representations (NeurIPS 2021 Competition) - Duration: 13 Dec 2021 → 14 Dec 2021 https://proceedings.mlr.press/v166/ |
Conference
Conference | HEAR: Holistic Evaluation of Audio Representations (NeurIPS 2021 Competition) |
---|---|
Period | 13/12/2021 → 14/12/2021 |
Internet address |