论文信息 - Tightly coupling Speech Recognition and Search

Tightly coupling Speech Recognition and Search

In this paper, we discuss the benefits of tightly coupling speech recognition and search components in the context of a speech-driven search application. We demonstrate that by incorporating constraints from the information repository that is being searched not only improves the speech recognition accuracy but also results in higher search accuracy.

Srinivas Bangalore | Taniya Mishra

[1] Otis Gospodnetic,et al. Lucene in Action (In Action series) , 2004 .

[2] K. Sparck Jones,et al. Simple, proven approaches to text retrieval , 1994 .

[3] Vladimir I. Levenshtein,et al. Binary codes capable of correcting deletions, insertions, and reversals , 1965 .

[4] Johan Schalkwyk,et al. Deploying GOOG-411: Early lessons in data, measurement, and testing , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.

[5] Geoffrey Zweig,et al. Live search for mobile:Web services by voice on the cellphone , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.

[6] K. Järvelin,et al. The RATF formula (Kwok's formula): exploiting average term frequency in cross-language retrieval , 2002, Inf. Res..

[7] Stephen E. Robertson,et al. Understanding inverse document frequency: on theoretical arguments for IDF , 2004, J. Documentation.