Automatic Speech Recognition with Very Large Conversational Finnish and Estonian Vocabularies
This tackles the problem of limited usability in speech recognition for conversational Finnish and Estonian, but appears incremental as it focuses on scaling vocabulary size rather than introducing a new paradigm.
The paper addresses the challenge of out-of-vocabulary words in automatic speech recognition for agglutinative languages like Finnish and Estonian, proposing the use of very large vocabularies with millions of word forms to cover colloquial pronunciations, compounding, and inflections, though no specific results or numbers are provided.
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.