End-to-end spoken language understanding without full transcripts

US11929062B2 · kind B2 · utility

Assignee

• International Business Machines Corporation · US

Inventors

• Hong-Kwang Kuo · Pleasantville, US
• Zoltan Tueske · White Plains, US
• Samuel Thomas · White Plains, US
• Yinghui Huang · New York, US
• Brian E. D. Kingsbury · Whitehall Corners, US
• Kartik Audhkhasi · White Plains, US

Key dates

• Filing date: Sep 15, 2020
• Grant date: Mar 12, 2024
• Expiry date: Jan 30, 2041

Classification

• Technology area (CPC G): Physics
• CPC primary: G06N3/048
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

A method and system of training a spoken language understanding (SLU) model includes receiving natural language training data comprising (i) one or more speech recording, and (ii) a set of semantic entities and/or intents for each corresponding speech recording. For each speech recording, one or more entity labels and corresponding values, and one or more intent labels are extracted from the corresponding semantic entities and/or overall intent. A spoken language understanding (SLU) model is trained based upon the one or more entity labels and corresponding values, and one or more intent labels of the corresponding speech recordings without a need for a transcript of the corresponding speech recording.