First European Conference on Speech Communication and Technology

Paris, France
September 27-29, 1989

Automatic Vocabulary Extension for a Speaker-Adaptive Speech Recognition System based on CVC Units

Peter Fesseler, Heidi Hackbarth, Marianne Kugler, Arnd Boehm

Standard Elektrik Lorenz AG, SEL Research Center, Stuttgart, F.R. Germany

For speech recognition with large vocabularies, a user should not be burdened with having to train several thousand words explicitly. Therefore, it proves extremely useful to provide a means for easy vocabulary generation and enlargement from written text input. Applying a set of appropriately defined rules, the orthography of a lexicon item is first transcribed into the phonetic symbols of the standard pronunciation and the most common alternatives thereof. From these, the multiple sequence of specific subword units of a lexicon entry is produced. The tool introduced here is part of a comprehensive speech processing system for subword-unit based, speaker-adaptive recognition of continuous speech.

Full Paper

Bibliographic reference.  Fesseler, Peter / Hackbarth, Heidi / Kugler, Marianne / Boehm, Arnd (1989): "Automatic vocabulary extension for a speaker-adaptive speech recognition system based on CVC units", In EUROSPEECH-1989, 1075-1078.