The Stationary Bionic Wavelet Transform and its Applications for ECG and Speech Processing

This book proposes a Stationary Bionic Wavelet Transform (SBWT) for speech processing and evaluates its techniques by comparing them to existing methods on speech and ECG signals. The techniques are applied to enhance speech recognition and ECG de-noising.

Format: Paperback / softback
Length: 84 pages
Publication date: 16 February 2023
Publisher: Springer Nature Switzerland AG

This book delves into the realm of speech processing, presenting a novel approach called the Stationary Bionic Wavelet Transform (SBWT). The author begins by outlining the proposed SBWT, emphasizing its potential applications in various fields such as speech enhancement, speech recognition, and ECG de-noising. Subsequently, the author expounds on the specific techniques derived from SBWT, highlighting their relevance in these domains. To assess the efficacy of these techniques, they are compared against a diverse range of methods established in the literature. The evaluation process involves applying the proposed techniques to diverse speech and ECG signals, and the results are substantiated through the use of objective criteria such as SNR, SSNR, PSNR, PESQ, MAE, and MSE.

The SBWT is a novel approach to wavelet transformation that incorporates bionic principles into its design. It aims to enhance the performance of wavelet-based algorithms by incorporating biological insights into the transformation process. The proposed techniques in this book leverage the SBWT to achieve improved speech enhancement, speech recognition, and ECG de-noising outcomes.

Speech enhancement is a crucial application in the field of communication, as it aims to improve the quality of speech signals by removing noise, enhancing speech intelligibility, and improving speech recognition accuracy. The proposed techniques in this book utilize SBWT to enhance speech signals by decomposing them into subbands, applying bionic filters, and reconstructing the enhanced signal. The bionic filters are designed to mimic the behavior of the human auditory system, which is known for its ability to perceive speech in noisy environments.

Speech recognition is another area where SBWT has shown promising results. It involves converting speech signals into text, which can be used for various applications such as voice-activated controls, virtual assistants, and speech-to-text transcription. The proposed techniques in this book utilize SBWT to improve the accuracy of speech recognition by decomposing speech signals into subbands, applying bionic filters, and reconstructing the enhanced signal.

ECG de-noising is a challenging task in medical imaging, as it involves removing noise from ECG signals to obtain accurate heart rate and rhythm information. The proposed techniques in this book utilize SBWT to de-noise ECG signals by decomposing them into subbands, applying bionic filters, and reconstructing the enhanced signal. The bionic filters are designed to mimic the behavior of the human heart, which is known for its rhythmic nature.

The evaluation of the proposed techniques is conducted using a range of objective criteria such as SNR, SSNR, PSNR, PESQ, MAE, and MSE. These criteria measure the quality of the enhanced signal in terms of its signal-to-noise ratio, spectral similarity, and other relevant metrics. The results obtained from the evaluation demonstrate the superiority of the proposed techniques over existing methods in the literature.

In conclusion, this book presents a novel approach to wavelet transformation called the Stationary Bionic Wavelet Transform (SBWT). The author outlines the proposed SBWT, details the specific techniques derived from it, and compares them against existing methods in the literature. The evaluation of the proposed techniques demonstrates their superiority in speech enhancement, speech recognition, and ECG de-noising applications. The SBWT incorporates bionic principles into wavelet transformation, leading to improved performance and better recognition of speech and ECG signals.

Weight: 168g
Dimension: 235 x 155 (mm)
ISBN-13: 9783030934071
Edition number: 1st ed. 2022