Language modeling for multi-domain speech-driven text retrieval
Tóm tắt
We report experimental results associated with speech-driven text retrieval, which facilitates retrieving information in multiple domains with spoken queries. Since users speak contents related to a target collection, we produce language models used for speech recognition based on the target collection, so as to improve both the recognition and retrieval accuracy. Experiments using existing test collections combined with dictated queries showed the effectiveness of our method.
Từ khóa
#Natural languages #Speech recognition #Information retrieval #Automatic speech recognition #Testing #Decoding #Libraries #Information science #Target recognition #Content based retrievalTài liệu tham khảo
lalit, 1983, A maximum linklihood approach to continuous speech recognition, IEEE Transactions on Pattern Analysis and Machine Intelligence, 5, 179
fabio, 2000, Word recognition errors and relevance feedback in spoken query processing, Proc of the International Conference on Flexible Query Answering Systems, 267
satoshi, 2000, IREX: IR and IE evaluation project in Japanese, Proc Int Conf on Lang Resources and Evaluation, 1475
itou, 0, The design of the newspaper-based Japanese large vocabulary continuous speech recognition corpus, ICSLP-98, 1998, 3261
kawahara, 2000, Free software toolkit for Japanese large vocabulary continuous speech recognition, Proc International Conference on Spoken Language Processing, 476
yuji, 1999, Japanese morphological analysis system ChaSen version 2.0 manual 2nd edition, Technical Report NAIST-IS-TR99008
10.1007/978-1-4471-2099-5_24
barnett, 0, Experiments in spoken queries for document retrieval, Proc EUROSPEECH, 1997, 1323
garofolo, 1997, TREC-6 1997 spoken document retrieval track overview and results, Proc 6th Text Retrieval Conf, 83
2001, National Institute of Informatics, Proceedings of the NTCIR Workshop 2 Meeting on Evaluation of Chinese and Japanese Text Retrieval and Text Summarization