Data sampling and dimensionality reduction approaches for reranking ASR outputs using discriminative language models


Dikici E., Semerci M., SARAÇLAR M., Alpaydin E.

12th Annual Conference of the International Speech Communication Association, INTERSPEECH 2011, Florence, İtalya, 27 - 31 Ağustos 2011, ss.1461-1464, (Tam Metin Bildiri)

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Florence
  • Basıldığı Ülke: İtalya
  • Sayfa Sayıları: ss.1461-1464
  • Anahtar Kelimeler: Data sampling, Dimensionality reduction, Discriminative language modeling, Ranking perceptron, Ranking SVM
  • Boğaziçi Üniversitesi Adresli: Evet

Özet

This paper investigates various approaches to data sampling and dimensionality reduction for discriminative language models (DLM). Being a feature based language modeling approach, the aim of DLM is to rerank the ASR output with discriminatively trained feature parameters. Using a Turkish morphology based feature set, we examine the use of online Principal Component Analysis (PCA) as a dimensionality reduction method. We exploit ranking perceptron and ranking SVM as two alternative discriminative modeling techniques, and apply data sampling to improve their efficiency. We obtain a reduction in word error rate (WER) of 0.4%, significant at p < 0:001 over the baseline perceptron result. Copyright © 2011 ISCA.