Invention Grant
- Patent Title: Minimum bayesian risk methods for automatic speech recognition
- Patent Title (中): 用于自动语音识别的最小贝叶斯风险方法
-
Application No.: US13771934Application Date: 2013-02-20
-
Publication No.: US09123333B2Publication Date: 2015-09-01
- Inventor: Antoine Amarilli , Mehryar Mohri , Cyril Allauzen
- Applicant: Google Inc.
- Applicant Address: US CA Mountain View
- Assignee: Google Inc.
- Current Assignee: Google Inc.
- Current Assignee Address: US CA Mountain View
- Agency: McDonnell Boehnen Hulbert & Berghoff LLP
- Main IPC: G10L15/26
- IPC: G10L15/26 ; G10L15/04 ; G10L15/18 ; G10L15/06 ; G10L15/08

Abstract:
A hypothesis space of a search graph may be determined. The hypothesis space may include n hypothesis-space transcriptions of an utterance, each selected from a search graph that includes t>n transcriptions of the utterance. An evidence space of the search graph may also be determined. The evidence space may include m evidence-space transcriptions of the utterance that are randomly selected from the search graph, where t>m. For each particular hypothesis-space transcription in the hypothesis space, an expected word error rate may be calculated by comparing the particular hypothesis-space transcription to each of the evidence-space transcriptions. Based on the expected word error rates, a lowest expected word error rate may be obtained, and the particular hypothesis-space transcription that is associated with the lowest expected word error rate may be provided.
Public/Granted literature
- US20140303973A1 Minimum Bayesian Risk Methods for Automatic Speech Recognition Public/Granted day:2014-10-09
Information query