Invention Grant
US07885812B2 Joint training of feature extraction and acoustic model parameters for speech recognition
有权
特征提取和声学模型参数的联合训练用于语音识别
- Patent Title: Joint training of feature extraction and acoustic model parameters for speech recognition
- Patent Title (中): 特征提取和声学模型参数的联合训练用于语音识别
-
Application No.: US11560180Application Date: 2006-11-15
-
Publication No.: US07885812B2Publication Date: 2011-02-08
- Inventor: Alejandro Acero , James G. Droppo , Milind V. Mahajan
- Applicant: Alejandro Acero , James G. Droppo , Milind V. Mahajan
- Applicant Address: US WA Redmond
- Assignee: Microsoft Corporation
- Current Assignee: Microsoft Corporation
- Current Assignee Address: US WA Redmond
- Agency: Westman, Champlin & Kelly, P.A.
- Agent Christopher J. Volkmann
- Main IPC: G10L15/06
- IPC: G10L15/06
Abstract:
Parameters for a feature extractor and acoustic model of a speech recognition module are trained. An objective function is utilized to determine values for the feature extractor parameters and the acoustic model parameters.
Public/Granted literature
- US20080114596A1 DISCRIMINATIVE TRAINING FOR SPEECH RECOGNITION Public/Granted day:2008-05-15
Information query
