Invention Grant
- Patent Title: System and method for a multiclass approach for confidence modeling in automatic speech recognition systems
-
Application No.: US16414885Application Date: 2019-05-17
-
Publication No.: US11195514B2Publication Date: 2021-12-07
- Inventor: Ramasubramanian Sundaram , Aravind Ganapathiraju , Yingyi Tan
- Applicant: Genesys Telecommunications Laboratories, Inc.
- Applicant Address: US CA Daly City
- Assignee: Genesys Telecommunications Laboratories, Inc.
- Current Assignee: Genesys Telecommunications Laboratories, Inc.
- Current Assignee Address: US CA Daly City
- Main IPC: G10L15/00
- IPC: G10L15/00 ; G10L15/10 ; G10L15/06 ; G06N20/00 ; G10L15/26 ; G10L15/14

Abstract:
A system and method are presented for a multiclass approach for confidence modeling in automatic speech recognition systems. A confidence model may be trained offline using supervised learning. A decoding module is utilized within the system that generates features for audio files in audio data. The features are used to generate a hypothesized segment of speech which is compared to a known segment of speech using edit distances. Comparisons are labeled from one of a plurality of output classes. The labels correspond to the degree to which speech is converted to text correctly or not. The trained confidence models can be applied in a variety of systems, including interactive voice response systems, keyword spotters, and open-ended dialog systems.
Public/Granted literature
Information query