System and method for enhancing low frequency spectrum content of a digitized voice signal
    41.
    发明授权
    System and method for enhancing low frequency spectrum content of a digitized voice signal 失效
    用于增强数字化语音信号的低频谱内容的系统和方法

    公开(公告)号:US06694291B2

    公开(公告)日:2004-02-17

    申请号:US09745244

    申请日:2000-12-20

    IPC分类号: G10L1906

    CPC分类号: G10L21/02 G10L21/0232

    摘要: A system for enhancing low frequency spectral content of a signal transmitted via a channel, that includes a noise suppression circuit to update channel energy estimates; and a spectral enhancer circuit coupled to and follows the noise suppression circuit, the spectral enhancer circuit to determine channel enhancement in response to the channel energy estimates.

    摘要翻译: 一种用于增强经由信道发送的信号的低频谱含量的系统,其包括用于更新信道能量估计的噪声抑制电路; 以及耦合到噪声抑制电路并且跟随噪声抑制电路的频谱增强器电路,频谱增强器电路响应于信道能量估计来确定信道增强。

    Noise-compensated speech recognition templates
    42.
    发明授权
    Noise-compensated speech recognition templates 失效
    噪声补偿语音识别模板

    公开(公告)号:US06381569B1

    公开(公告)日:2002-04-30

    申请号:US09018257

    申请日:1998-02-04

    IPC分类号: G10L1520

    CPC分类号: G10L15/20 G10L21/0216

    摘要: The speech recognition training unit is modified to store digitized speech samples into a speech database that can be accessed at recognition time. The improved recognition unit comprises a noise analysis, modeling, and synthesis unit which continually analyzes the noise characteristics present in the audio environment and produces an estimated noise signal with similar characteristics. The recognition unit then constructs a noise-compensated template database by adding the estimated noise signal to each of the speech samples in the speech database and performing parameter determination on the resulting sums. This procedure accounts for the presence of noise in the recognition phase by retraining all the templates using an estimated noise signal with similar characteristics as the actual noise signal that corrupted the word to be recognized. This method improves the likelihood of a good template match, which increases the recognition accuracy.

    摘要翻译: 修改语音识别训练单元以将数字化语音样本存储到可在识别时被访问的语音数据库中。 改进的识别单元包括噪声分析,建模和合成单元,其连续分析存在于音频环境中的噪声特性并产生具有相似特性的估计噪声信号。 然后,识别单元通过将估计的噪声信号加到语音数据库中的每个语音样本上并对所得到的和进行参数确定来构建噪声补偿模板数据库。 该过程通过使用具有与损坏要识别的字的实际噪声信号相似的特性的估计噪声信号重新训练所有模板来解决识别阶段中的噪声的存在。 该方法提高了模板匹配的可能性,从而提高了识别精度。

    Multi-user wireless telephone having dual echo cancellers
    43.
    发明授权
    Multi-user wireless telephone having dual echo cancellers 失效
    具有双重回声消除器的多用户无线电话

    公开(公告)号:US5999828A

    公开(公告)日:1999-12-07

    申请号:US820934

    申请日:1997-03-19

    CPC分类号: H04M9/082 H04B3/23

    摘要: A method and apparatus for canceling both earseal and hybrid echo in a Wireless Local Loop telephone system is disclosed. The present invention operates within a subscriber station having a plurality of telephone inputs, one of which is a dedicated handset and at least one other input being a standard analog telephone. During a conference call between a far end speaker and two near end speakers, the far end speaker may hear an echo of his own voice due to an earseal echo generated by the dedicated handset and a hybrid echo generated by a 4-to-2 wire hybrid interface within the subscriber station. In accordance with the present invention, two distinct echo cancellers are used in the subscriber station to reduce the two types of echo. A first echo canceller is optimized to remove the earseal echo generated at the dedicated handset while a second echo canceller is optimized to remove the hybrid echo generated at the 4-to-2 wire hybrid. The two echo cancellers operate independently such that neither echo canceller need adapt its filter coefficients when a change in the echo channel characteristics occur, such as when a speaker hangs up or another speaker joins the conversation.

    摘要翻译: 公开了一种在无线本地环路电话系统中消除耳塞和混合回波的方法和装置。 本发明在具有多个电话输入的用户站中操作,其中一个是专用手持机,并且至少一个其它输入是标准模拟电话。 在远端扬声器和两个近端扬声器之间的电话会议期间,远端扬声器可能由于专用手机产生的耳塞回声和由4对2线产生的混合回声而听到自己的声音的回声 用户站内的混合接口。 根据本发明,在用户站中使用两个不同的回波消除器来减少两种类型的回波。 优化第一回波消除器以消除在专用手持机处产生的耳塞回声,而第二回波消除器被优化以去除在4对2线混合电路产生的混合回波。 两个回波消除器独立地操作,使得当回波信道特性发生变化时,例如当扬声器挂起或者另一个扬声器加入对话时,回波消除器都不需要调整其滤波器系数。