Semi-supervised methods for improving keyword search of unseen terms
Scott Novotney, Ivan Bulyko, Richard Schwartz, Sanjeev P. Khudanpur, Owen Kimball · 2012
We present a semi-supervised language modeling technique to improve search performance on words without training data. Probabilities estimated from automatic transcripts of a large corpus of in-domain audio are added to an existing LM. Requiring only pronunciations and indomain audio, our method achieves 70% of the possible gain without any supervised data or development data. This method is contrasted with techniques using additional resources. If human effort is available, we also present a transcription regime to efficiently close the gap.