Wan, Chia-YuChia-YuWanLIN-SHAN LEE2020-06-112020-06-11200815587916https://scholars.lib.ntu.edu.tw/handle/123456789/498541https://www.scopus.com/inward/record.uri?eid=2-s2.0-64849105676&doi=10.1109%2fTASL.2008.920891&partnerID=40&md5=6d8c5e68d1be273669883111b7cd46d2In a distributed speech recognition (DSR) framework, the speech features are quantized and compressed at the client and recognized at the server. However, recognition accuracy is degraded by environmental noise at the input, quantization distortion, and transmission errors. In this paper, histogram-based quantization (HQ) is proposed, in which the partition cells for quantization are dynamically defined by the histogram or order statistics of a segment of the most recent past values of the parameter to be quantized. This scheme is shown to be able to solve to a good degree many problems related to DSR. A joint uncertainty decoding (JUD) approach is further developed to consider the uncertainty caused by both environmental noise and quantization errors. A three-stage error concealment (EC) framework is also developed to handle transmission errors. The proposed HQ is shown to be an attractive feature transformation approach for robust speech recognition outside of a DSR environment as well. All the claims have been verified by experiments using the Aurora 2 testing environment, and significant performance improvements for both robust and/or distributed speech recognition over conventional approaches have been achieved. © 2008 IEEE.Error compensation; Robustness; Speech recognition; Vector quantization (VQ)[SDGs]SDG16Conventional approaches; Distributed speech recognition; Environmental noise; Error concealments; Feature transformations; Joint uncertainties; Order statistics; Partition cells; Performance improvements; Quantization distortions; Quantization errors; Recognition accuracies; Robust speech recognition; Robustness; Speech features; Testing environments; Transmission errors; Decoding; Error compensation; Error detection; Speech analysis; Vector quantization; Speech recognitionjournal article10.1109/TASL.2008.9208912-s2.0-64849105676WOS:000258033600016