Search International and National Patent Collections

1. (WO2018161763) TRAINING METHOD FOR VOICE DATA SET, COMPUTER DEVICE AND COMPUTER READABLE STORAGE MEDIUM

Pub. No.:    WO/2018/161763    International Application No.:    PCT/CN2018/075595
Publication Date: Fri Sep 14 01:59:59 CEST 2018 International Filing Date: Thu Feb 08 00:59:59 CET 2018
IPC: G10L 15/14
Applicants: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
腾讯科技(深圳)有限公司
Inventors: SUN, Tao
孙涛
KANG, Yueteng
康跃腾
ZHANG, Xiaoming
张晓明
ZHANG, Li
张力
Title: TRAINING METHOD FOR VOICE DATA SET, COMPUTER DEVICE AND COMPUTER READABLE STORAGE MEDIUM
Abstract:
A training method for a voice data set, comprising: reading a first test set generated by selecting data from a first voice data set, and acquiring a first voice model parameter obtained by training the first voice data set (202); acquiring a second voice data set, and randomly selecting data from the second voice data set to generate a second test set (204); and when detecting that the second test set and the first test set meet a similar condition, performing second voice model training on the second voice data set using the first voice model parameter obtained by training (206).