Multi-Attention Bottleneck for Gated Convolutional Encoder-Decoder-Based Speech Enhancement

• Published in: IEEE access Vol. 11; pp. 114172 - 114186
• Main Authors: Saleem, Nasir, Gunawan, Teddy Surya, Shafi, Muhammad, Bourouis, Sami, Trigui, Aymen
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; channel attention; Computer architecture; Convolutional neural networks; Decoding; Demand; Encoders-Decoders; Encoding; gated convolutional encoder-decoder; Intelligibility; Logic gates; Modules; Multi-attention; Neural networks; Noise measurement; Speech; Speech enhancement; Speech processing; Time-frequency analysis; time-frequency attention; transformer; Transformers

Convolutional encoder-decoder (CED) has emerged as a powerful architecture, particularly in speech enhancement (SE), which aims to improve the intelligibility and quality and intelligibility of noise-contaminated speech. This architecture leverages the strength of the convolutional neural networks (...