Confidence Measures for Spontaneous Speech Recognition
ICASSP '97 Proceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP '97)-Volume 2 - Volume 2
A maximum entropy approach to named entity recognition
A maximum entropy approach to named entity recognition
Japanese named entity extraction evaluation: analysis of results
COLING '00 Proceedings of the 18th conference on Computational linguistics - Volume 2
Improving information extraction by modeling errors in speech recognizer output
HLT '01 Proceedings of the first international conference on Human language technology research
Efficient support vector classifiers for named entity recognition
COLING '02 Proceedings of the 19th international conference on Computational linguistics - Volume 1
Named entity recognition with a maximum entropy approach
CONLL '03 Proceedings of the seventh conference on Natural language learning at HLT-NAACL 2003 - Volume 4
CONLL '03 Proceedings of the seventh conference on Natural language learning at HLT-NAACL 2003 - Volume 4
Robust named entity extraction from large spoken archives
HLT '05 Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing
Using N-best lists for named entity recognition from Chinese speech
HLT-NAACL-Short '04 Proceedings of HLT-NAACL 2004: Short Papers
Hi-index | 0.00 |
This paper proposes a named entity recognition (NER) method for speech recognition results that uses confidence on automatic speech recognition (ASR) as a feature. The ASR confidence feature indicates whether each word has been correctly recognized. The NER model is trained using ASR results with named entity (NE) labels as well as the corresponding transcriptions with NE labels. In experiments using support vector machines (SVMs) and speech data from Japanese newspaper articles, the proposed method outperformed a simple application of text-based NER to ASR results in NER F-measure by improving precision. These results show that the proposed method is effective in NER for noisy inputs.