Two-stage vocabulary-free spoken document retrieval - subword identification and re-recognition of the identified sections.
Yoshiaki ItohTakayuki OtakeKohei IwataKazunori KojimaMasaaki IshigameKazuyo TanakaShi-wook LeePublished in: INTERSPEECH (2006)
Keyphrases
- spoken document retrieval
- out of vocabulary
- speech corpus
- information retrieval
- broadcast news
- cross language
- spoken documents
- test collection
- feature extraction
- language model
- question answering
- cross language information retrieval
- n gram
- pattern recognition
- cross lingual
- retrieval method
- parallel corpora
- named entity recognition
- word segmentation
- unsupervised learning
- multi modal
- multimedia
- metadata