首页> 外国专利> The speech input system which accompanies standard conversion expedient

The speech input system which accompanies standard conversion expedient

机译:符合标准转换权的语音输入系统

摘要

(57) Abstract The audio information (SI) which is stained by the respective reception channel (21,25,28), the 1st reception channel (21) and through the 2nd reception channel (25,28), being the speech input system (8) which is given, the standard accumulation expedient (36) which accumulates the standard information (RI1) which features the type of pronunciation of the word by the group of standard speakers and the standard information (RI and ARI) which is accumulated, 1st it is used by the user or it is adapted to the 2nd reception channel (21,25,28), reception channel adaptation expedient (30,38,44) and, it was accumulatedStandard information (RI1, RI2 and RI3), it is adapted to the type of pronunciation of the word by the user of the speech input system (8), to possess with user adaptation expedient (37), at the same time, to possess the speech recognition expedient (29) which recognizes the text information (TI) which it should you allot to the audio information (SI) which is supplied, the standard information (ARI1, ARI2 and ARI3) which is adapted on the one hand, by reception channel adaptation expedient (30,38,44) and user adaptation expedient (37) to be appraised, as for reception channel adaptation expedient (30,38,44), the conversion matrix (T1-2, T1-3 and T2-3)Following, in order the 1st standard information (RI1 and ARI1) which is adapted to the 1st reception channel (21) to the 2nd standard information (RI2, RI3, ARI2 and ARI3) which is adapted to the 2nd reception channel (25,28) to convert, it includes the standard conversion expedient (T1-2, T1-3 and T2-3) which you can do, is converted the 1st standard information (RI1 and ARI1) which is adapted, already, but necessity is not that, on the one hand, depending upon standard conversion expedient (T1-2, T1-3 and T2-3), but being adapted by the user by user adaptation expedient (37)The speech input system which is.
机译:(57)<摘要>被相应的接收通道(21,25,28),第一接收通道(21)以及通过第二接收通道(25,28)染色的音频信息(SI)是语音在给定的输入系统(8)中,标准累积权宜(36)累积标准信息(RI1),该信息具有一组标准说话人的单词发音类型,而标准信息(RI和ARI)为首先,它被用户使用,或者它被适配到第二个接收信道(21,25,28),接收信道适配权(30,38,44),并且被累积了标准信息(RI1,RI2和RI3) ,它适用于语音输入系统(8)的用户的单词发音类型,以具有用户适应权宜之物(37),同时具有语音识别权宜之物(29)您应该分配给s的音频信息(SI)的文本信息(TI)一方面,通过评估接收信道自适应权宜(30,38,44)和用户自适应权宜(37)一方面调整标准信息(ARI1,ARI2和ARI3)在图30、38、44)中,转换矩阵(T1-2,T1-3和T2-3)依次遵循适用于第一接收信道(21)至第二接收的第一标准信息(RI1和ARI1)。适用于要转换的第二接收通道(25,28)的标准信息(RI2,RI3,ARI2和ARI3),其中包括您可以执行的标准转换权宜之计(T1-2,T1-3和T2-3)被转换的第一标准信息(RI1和ARI1)已经被适配,但是一方面不是必须的,这取决于标准转换权宜之计(T1-2,T1-3和T2-3),但是(37)语音输入系统。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号