No abstract

This paper investigates the error mitigation algorithms for distributed speech recognition over wireless channels. A MAP symbol decoding algorithm which exploits the combined a priori information of source and channel is proposed. This is used in conjunction with a modified BCJR algorithm for decoding convolutional codes based on sectionalized code trellises. Performance is further enhanced by the use of the Gilbert channel model that more closely characterizes the statistical dependencies between channel bit errors. Experiments on Mandarin digit string recognition task indicate that our proposed mitigation scheme achieves high robustness against channel errors.

Packet loss and delay are two essential problems to realtime voice transmission over IP networks. In the proposed system, multiple descriptions of the speech are transmitted to take advantage of largely uncorrelated delay and loss characteristics on different network paths. Adaptive playout scheduling of multiple voice streams is formulated as an optimization problem leading to a better delay-loss tradeoff. Also proposed is a perceptually motivated optimization criterion based on a simplified version of the ITU-T E-model. Experimental results show that the proposed multi-stream playout algorithm improves the delay-loss tradeoff as well as speech reconstruction quality.

No abstract