公开(公告)号：WO2015003596A1

公开(公告)日：2015-01-15

申请号：PCT/CN2014/081763

申请日：2014-07-07

Applicant: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED

Inventor： CHU, Pan ,  XU, Hao ,  WANG, Fei ,  HAN, Guiping ,  LIAO, Ling

IPC: G06F17/30

CPC classification number: G10L15/06 ,  G10L17/22 ,  G10L19/173 ,  G10L21/0364 ,  G10L21/04 ,  H04L51/12 ,  H04L51/28

Abstract: Systems,devices and methods are provided for configuring matching rules related to voice input commands. For example,a first mapping relation between one or more first original terms in a preset term database and one or more first identification terms is established; the first mapping relation is stored in a first mapping relation table; one or more first voice input commands are configured for the first identification terms or one or more first statements including the first identification terms; and a second mapping relation between the first identification terms or the first statements and the first voice input commands is stored into a second mapping relation table.

Abstract translation: 提供了系统，设备和方法来配置与语音输入命令相关的匹配规则。 例如，建立预设术语数据库中的一个或多个第一原始术语与一个或多个第一识别项之间的第一映射关系; 第一映射关系被存储在第一映射关系表中; 一个或多个第一语音输入命令被配置用于第一识别项或包括第一识别项的一个或多个第一语句; 并且第一识别项或第一语句和第一语音输入命令之间的第二映射关系被存储到第二映射关系表中。

公开(公告)号：WO2014201834A1

公开(公告)日：2014-12-24

申请号：PCT/CN2013/090468

申请日：2013-12-25

Applicant: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED

Inventor： HAN, Guiping ,  XU, Hao ,  ZHU, Pan ,  WANG, Fei ,  LIAO, Ling

IPC: G10L15/04

CPC classification number: G10L15/26 ,  G06F17/2223

Abstract: A method and device for matching speech to text are disclosed, the method including: receiving a speech input, the mentioned speech input carrying input speech information; obtaining initial text corresponding to the input speech information, and respective pinyin of the initial text; generating at least one approximate pinyin for the initial text based on predetermined pronunciation similarity information; and from a preset mapping relationship table, obtaining additional text corresponding to the respective pinyin of the initial text or to the at least one approximate pinyin of the initial text, wherein the preset mapping relationship table includes a respective record for each word in a word database, including respective pinyin and at least one respective approximate pinyin for said each word, and a respective mapping relation between said respective pinyin, said at least one respective approximate pinyin, and said each word.

Abstract translation: 公开了一种用于将语音与文本进行匹配的方法和装置，所述方法包括：接收语音输入，所述提及的输入语音信息的语音输入; 获取对应于输入语音信息的初始文本，以及初始文本的各自拼音; 基于预定的发音相似性信息，产生用于初始文本的至少一个近似拼音; 并且从预设映射关系表中获取对应于初始文本的相应拼音的附加文本或初始文本的至少一个近似拼音，其中预设映射关系表包括单词数据库中的每个单词的相应记录 ，包括针对所述每个单词的各自拼音和至少一个相应的近似拼音，以及所述相应拼音，所述至少一个相应的近似拼音和所述每个单词之间的相应映射关系。