Publication Details
Recurrent Neural Network Language Modeling Applied to the Brno AMI/AMIDA 2009 Meeting Recognizer Setup
automatic speech recognition, language modeling, recurrent neural networks
This paper is on Recurrent Neural Network Language Modeling Applied to the Brno AMI/AMIDA 2009 Meeting Recognizer Setup.
In this paper we use recurrent neural network (RNN) based language models to improve our 2009 English meeting recognizer originated from the AMI/AMIDA project, which to date was the most advanced speech recognition setup of the Speech@FIT. On the baseline setup using the original language models we decrease word error rate (WER) from 20.3% to 19.1%. When language models in the system are replaced by models trained on a tiny subset of the original language model data, WER drops from 22.2% to 20.4%. Adding data sampled from two RNN models for language model training improves the overall system, yielding the performance of the original baseline (20.2%).
@INPROCEEDINGS{FITPUB9691,
author = "Stefan Kombrink and Tom\'{a}\v{s} Mikolov",
title = "Recurrent Neural Network Language Modeling Applied to the Brno AMI/AMIDA 2009 Meeting Recognizer Setup",
pages = "527--531",
booktitle = "Proceedings of the 17th Conference STUDENT EEICT 2011",
series = "Volume 3",
year = 2011,
location = "Brno, CZ",
publisher = "Brno University of Technology",
ISBN = "978-80-214-4273-3",
language = "english",
url = "https://www.fit.vut.cz/research/publication/9691"
}