Automatic Speech Recognition with Very Large Conversational Finnish and Estonian Vocabularies

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.

References in corpus (7)

• Efficient Estimation of Word Representations in Vector Space
• Distributed Representations of Words and Phrases and their Compositionality
• Training Very Deep Networks
• A Fast and Simple Algorithm for Training Neural Probabilistic Language Models
• Efficient softmax approximation for GPUs
• BlackOut: Speeding up Recurrent Neural Network Language Models With Very Large Vocabularies
• TheanoLM - An Extensible Toolkit for Neural Network Language Modeling

Cited by in corpus (4)

• Investigation on N-gram Approximated RNNLMs for Recognition of Morphologically Rich Speech
• On the Effectiveness of Neural Text Generation based Data Augmentation for Recognition of Morphologically Rich Speech
• Effects of Language Relatedness for Cross-lingual Transfer Learning in Character-Based Language Models
• The Current State of Finnish NLP