SMS based interface for FAQ retrieval
ACL '09 Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP: Volume 2 - Volume 2
Handling noisy queries in cross language FAQ retrieval
EMNLP '10 Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing
AND '10 Proceedings of the fourth workshop on Analytics for noisy unstructured text data
Hi-index | 0.00 |
An application which provides a service of producing speech for any text entered by the user is highly beneficial for the society. A person with speaking disabilities would benefit the most from it. In this paper we have proposed and developed a software system that performs noise reduction and text-to-speech conversion. For removal of noise, text entered by the user is checked for any misspelling or non-standard abbreviation. Each word in the text is processed as a separate token. Each token is checked using three available algorithms - Longest Common Subsequence, Levenshtein distance and Soundex algorithms. Using the above mentioned algorithms we select the best possible variant of the word from a pre-defined wordlist and we replace the word with the best possible replacement. After noise reduction, text is sent as input to a text-to-speech convertor which produces the corresponding sound file which can be played and is audible to the user. This sound file also could be saved for future use.