Vocabulary and/or language model training

US6430551B1 · kind B1 · utility

Assignee

• KONINKLIJKE PHILIPS ELECTRONICS NV · NL

Inventors

• Eric Thelen · Aachen, DE
• Stefan Besling · Aachen, DE
• Steven DeJarnett · San Jose, US

Key dates

• Filing date: Oct 6, 1998
• Grant date: Aug 6, 2002
• Expiry date: Oct 6, 2018

Classification

• Technology area (CPC Y): Emerging Cross-Sectional Technologies
• CPC primary: Y10S707/99936
• WIPO field: Control
• WIPO sector: Instruments

Abstract

A system includes means for creating a vocabulary and/or statistical language model from a textual training corpus. The vocabulary and/or language model are used in a pattern recognition system, such as a speech recognition system or a handwriting recognition system, for recognizing a time-sequential input pattern. The system includes means for determining at least one context identifier and means for deriving at least one search criterion, such as a keyword, from the context identifier. The system further includes means for selecting documents from a set of documents based on the search criterion. Advantageously, an Internet search engine is used for selecting the documents. Means are used for composing the training corpus from the selected documents.