基於多感測器之生醫訊號收集及加密演算法開發
Multi-sensors-based biomedical data collection and encryption
For speech-related applications in Internet of things environments, identifying effective methods to handle interference noises and compress the amount of data in transmissions is essential for achieving high-quality services. We proposed a novel multi-input multi-output speech compression and enhancement (MIMO-SCE) system based on a convolutional denoising auto-encoder (CDAE) model to simultaneously improve speech quality and reduce the dimension of transmission data. Compared with conventional single-channel and multi-input single-output systems, MIMO systems can be employed for applications where multiple acoustic signals need to be handled. We investigated two CDAE models, fully convolutional network (FCN) and Sinc FCN, as the core models in MIMO systems. The experimental results confirm that the proposed MIMO-SCE framework effectively improves speech quality and intelligibility, while reducing the amount of recording data to one-seventh for transmission.
對於物聯網還境中語音相關的各種應用，能夠有效地抑制干擾噪聲及壓縮語音資料，對於實現高品質服務至關重要。基於此，我們提出了一套卷積去噪自動編碼器（CDAE）架構的多輸入多輸出語音壓縮和增強系統（MIMO-SCE），此系統可以同時提高語音質量及減小傳輸數據量。與傳統的單通道和多輸入單輸出系統相比，MIMO系統可用於需要處理多個聲音信號的應用。我們研究了兩個CDAE模型，即全卷積網路（FCN）和Sinc FCN，作為MIMO系統的核心架構。實驗結果證實，提出的MIMO-SCE框架有效地提高了語音質量和清晰度，同時將語音資料的總量壓縮至了七分之一。
