ISCA Archive SLTU 2014
ISCA Archive SLTU 2014

Rescoring n-best lists for Russian speech recognition using factored language models

Irina Kipyatkova, Vasilisa Verkhodanova, Alexey Karpov

In this paper, we present a research of factored language model (FLM) for rescoring N-best lists for Russian speech recognition task. As a baseline language model we used a 3- gram language model. Both baseline and factored language models were trained on a text corpus collected from recent news texts on Internet sites of online newspapers; total size of the corpus is about 350 million words (2.4 GB data). For FLMs creation, we used five factors: word, its lemma, stem, part-of-speech, and morphological tag. We investigate the influence of factor set on language model perplexity and word error rate (WER). Experiments on large vocabulary continuous Russian speech recognition showed that FLM can reduce WER.

Index Terms: factored language model (FLM), automatic speech recognition (ASR), N-best lists, Russian language processing


Cite as: Kipyatkova, I., Verkhodanova, V., Karpov, A. (2014) Rescoring n-best lists for Russian speech recognition using factored language models. Proc. 4th Workshop on Spoken Language Technologies for Under-Resourced Languages (SLTU 2014), 81-86

@inproceedings{kipyatkova14_sltu,
  author={Irina Kipyatkova and Vasilisa Verkhodanova and Alexey Karpov},
  title={{Rescoring n-best lists for Russian speech recognition using factored language models}},
  year=2014,
  booktitle={Proc. 4th Workshop on Spoken Language Technologies for Under-Resourced Languages  (SLTU 2014)},
  pages={81--86}
}