A method for correcting errors in speech recognition using the statistical features of character co-occurrence
- 1 January 1998
- conference paper
- Published by Association for Computational Linguistics (ACL)
- Vol. 1, 653
- https://doi.org/10.3115/980451.980954