This paper presents speech-driven Web retrieval models which accept spoken search topics (queries) in the NTCIR-3 Web retrieval task. The major focus of this paper is on improving speech recognition accuracy of spoken queries and then improving retrieval accuracy in speech-driven Web retrieval. We experimentally evaluated the techniques of combining outputs of multiple LVCSR models in recognition of spoken queries. As model combination techniques, we compared the SVM learning technique with conventional voting schemes such as ROVER. In addition, for investigating the effects on the retrieval performance in vocabulary size of the language model, we prepared two kinds of language models: the one's vocabulary size was 20,000, the other's one was 60,000. Then, we evaluated the differences in the recognition rates of the spoken queries and the retrieval performance. We showed that the techniques of multiple LVCSR model combination could achieve improvement both in speech recognition and retrieval accuracies in speech-driven text retrieval. Comparing with the retrieval accuracies when an LM with a 20,000/60,000 vocabulary size is used in an LVCSR system, we found that the larger the vocabulary size is, the better the retrieval accuracy is.

This paper describes an accurate unsupervised speaker adaptation method for lecture style spontaneous speech recognition using multiple LVCSR systems. In an unsupervised speaker adaptation framework, the improvement of recognition performance by adapting acoustic models remarkably depends on the accuracy of labels such as phonemes and syllables. Therefore, extraction of the adaptation data guided by confidence measure is effective for unsupervised adaptation. In this paper, we looked for the high confidence portions based on the agreement between two LVCSR systems, adapted acoustic models using the portions attached with high accurate labels, and then improved the recognition accuracy. We applied our method to the Corpus of Spontaneous Japanese (CSJ) and the method improved the recognition rate by about 2.1% in comparison with a traditional method.

This paper proposes and performs the primary feasibility evaluation on Flow Attribute Notification Protocol (FANP), which is a protocol between neighbor CSR (Cell Switch Router) nodes for the management of cut-through packet forwarding, in order to apply label switching paradigm. In cut-through packet forwarding with label switching, a router doesn't have to perform conventional IP packet processing for the received packets. FANP indicates the mapping information between a data-link connection and a packet flow to the neighbor node. FANP defines two key procedures, i. e. , one is the VCID Notification Procedure, and the other is the Flow-ID Notification Procedure. The VCID Notification Procedure lets the label switching paradigm over the label swapped data-link, such as ATM link, though the other label switch architecture can not work over the label swapped data-link. The primary evaluation of FANP has been performed using the prototype system and with the actual packet statistics. The result shows that, with a corporate backbone level, the label switch router system with FANP would work well.

Overview of Cell Switch Router (CSR) and the CSR prototype system are described. CSR can simultaneously support both connection oriented IP flows and connectionless IP flows. CSR contains cell switch fabric and IP packet switch fabric to achieve high throughput IP forwarding. IP packets are forwarded either through a cut-thru packet transmission, in which packet are forwarded without reassembling IP packet nor IP header processing, or through a conventional hop-by-hop IP packet forwarding. This paper describes and proposes the mechanism to forward the connectionless IP packet flows at the CSR. A CSR prototype system has been developed. The CSR prototype system uses PVC connections to transfer the IP packets. With the CSR prototype system, we can make sure that CSR system can achieve a high throughput, i.e., 2.4 Gbps aggregated throughput. For end-to-end TCP/IP packet transmission, more than 90 Mbps can be achieved and realtime video transmission (30 Mbps video) can be achieved.