Automatic Speech Recognition With Very Large Conversational Finnish and Estonian Vocabularies
Authors:
- 23 August 2017
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 25 (11) , 2085-2097
Abstract
Today, the vocabulary size for language models in large vocabulary speech recognition is typically several hundreds of thousands of words. While this is already sufficient in some applications, the out-of-vocabulary words are still limiting the usability in others. In agglutinative languages the vocabulary for conversational speech should include millions of word forms to cover the spelling variations due to colloquial pronunciations, in addition to the word compounding and inflections. Very large vocabularies are also needed, for example, when the recognition of rare proper names is important. Previously, very large vocabularies have been efficiently modeled in conventional n-gram language models either by splitting words into subword units or by clustering words into classes. While vocabulary size is not as critical anymore in modern speech recognition systems, training time and memory consumption become an issue when state-of-the-art neural network language models are used. In this paper, we investigate techniques that address the vocabulary size issue by reducing the effective vocabulary size and by processing large vocabularies more efficiently. The experimental results in conversational Finnish and Estonian speech recognition indicate that properly defined word classes improve recognition accuracy. Subword n-gram models are not better on evaluation data than word n-gram models constructed from a vocabulary that includes all the words in the training corpus. However, when recurrent neural network (RNN) language models are used, their ability to utilize long contexts gives a larger gain to subword-based modeling. Our best results are from RNN language models that are based on statistical morphs. We show that the suitable size for a subword vocabulary depends on the language. Using time delay neural network acoustic models, we were able to achieve new state of the art in Finnish and Estonian conversational speech recognition, 27.1% word error rate in the Finnish task and 21.9% in the Estonian task.Keywords
All Related Versions
Funding Information
- Research Council of Finland (251170, 274075)
- Kone Foundation
This publication has 26 references indexed in Scilit:
- Improved Subword Modeling for WFST-Based Speech Recognition2017
- Low-rank matrix factorization for Deep Neural Network training with high-dimensional output targets2013
- Morph-based speech recognition and modeling of out-of-vocabulary words across languagesACM Transactions on Speech and Language Processing, 2007
- On Growing and Pruning Kneser–Ney Smoothed $ N$-Gram ModelsIEEE Transactions on Audio, Speech, and Language Processing, 2007
- Unsupervised models for morpheme segmentation and morphology learningACM Transactions on Speech and Language Processing, 2007
- Unlimited vocabulary speech recognition with morph language models applied to FinnishComputer Speech & Language, 2006
- Generalized algorithms for constructing statistical language models2003
- Unsupervised discovery of morphemes2002
- Normalization of non-standard wordsComputer Speech & Language, 2001
- Long Short-Term MemoryNeural Computation, 1997