Details
Citation
Raza M, Ahmed K, Muzaffar J & Adeel A (2023) Two-Point Neurons for Efficient Multimodal Speech Enhancement. In: 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW), Rhodes Island, Greece, 04.06.2023-10.06.2023. IEEE, pp. 1-5. https://doi.org/10.1109/icasspw59220.2023.10193457
Abstract
Here we present a two-point neuron-inspired deep convolutional net (DCN) with 18 convolutional layers for multimodal speech enhancement (MM-SE) and compare it against conventional point neuron-inspired DCN in terms of Perceptual Evaluation of Speech Quality (PESQ) and Short-Time Objective Intelligibility (STOI). We show that the two-point neuron-driven DCN performs comparably to point-neurons driven DCN by using only ≈0.2% neurons at any time during training.
| Status | Published |
|---|---|
| Funders | |
| Publication date | 30/06/2023 |
| Publication date online | 31/08/2023 |
| Publisher | IEEE |
| Conference | 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW) |
| Conference location | Rhodes Island, Greece |
| Dates |
People (1)
Assoc. Prof. in Artificial Intelligence, Computing Science and Mathematics - Division