OOV Proper Name retrieval using topic and lexical context models

Retrieving Proper Names (PNs) specific to an audio document can be useful for vocabulary selection and OOV recovery in speech recognition, as well as in keyword spotting and audio indexing tasks. We propose methods to infer and retrieve OOV PNs relevant to an audio news document by using probabilist...

Full description

Saved in:
Bibliographic Details
Published in2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) pp. 5291 - 5295
Main Authors Sheikh, Imran, Illina, Irina, Fohr, Dominique, Linares, Georges
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.04.2015
Subjects
Context
Context modeling
Mathematical model
OOV
Probabilistic logic
proper names
Speech
Speech recognition
Vocabulary
Online AccessGet full text

Cover

More Information
Summary:Retrieving Proper Names (PNs) specific to an audio document can be useful for vocabulary selection and OOV recovery in speech recognition, as well as in keyword spotting and audio indexing tasks. We propose methods to infer and retrieve OOV PNs relevant to an audio news document by using probabilistic topic models trained over diachronic text news. LVCSR hypothesis on the audio news document is analysed for latent topics, which is then used to retrieve relevant OOV PNs. Using an LDA topic model we obtain Recall up to 0.87 and Mean Average Precision (MAP) of 0.26 with only top 10% of the retrieved OOV PNs. We further propose methods to re-score and retrieve rare OOV PNs, and a lexical context model to improve the target OOV PN rankings assigned by the topic model, which may be biased due to prominence of certain news events. Re-scoring rare OOV PNs improves Recall whereas the lexical context model improves MAP.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2015.7178981