利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

4 条记录 (耗时 0.01 秒)

    Method of speeding the execution of neural networks for correlated signal processing
    1.
    发明公开
    Method of speeding the execution of neural networks for correlated signal processing 失效
    Verfahren zur Beschleunigung derAusführungsgeschwindigkeitNeuronalnetzwerkenfürkorrelierte Signalverarbeitung

    公开(公告)号：EP0733982A1

    公开(公告)日：1996-09-25

    申请号：EP96104445.0

    申请日：1996-03-20

    申请人： CSELT Centro Studi e Laboratori Telecomunicazioni S.p.A.

    发明人： Albesano, Dario ,  Gemello, Roberto ,  Mana, Franco

    IPC分类号： G06F15/80

    CPC分类号： G06N3/0481 ,  G06N3/063

    摘要： The method allows speeding up the execution of a wide class of neural networks for processing input signals evolving slowly through time, such as, for instance, voice, radar, sonar, video signals, and it requires no specialised, costly or hard-to-find hardware. The method requires storing, for the neurons in at least one level of the network, the activation value at a certain instant and comparing it with the one computed at the subsequent instant. If the activation is equal, the neuron carries out no activity, otherwise it propagates the difference in activation, multiplied by the interconnection weights, to the neurons it is connected to.

    摘要翻译： 该方法允许加速执行广泛类型的神经网络，以处理通过时间缓慢进行的输入信号，例如语音，雷达，声纳，视频信号，并且不需要专门的，昂贵的或难以实现的， 找硬件。 该方法需要在网络的至少一个级别的神经元中存储在特定时刻的激活值并将其与在随后的时刻计算出的激活值进行比较。 如果激活相等，则神经元不进行任何活动，否则将激活差异乘以互连权重传播到其所连接的神经元。

    Speaker independent isolated word recognition system using neural networks
    4.
    发明公开
    Speaker independent isolated word recognition system using neural networks 失效
    用于使用神经网络孤立的单词说话者无关的识别系统。

    公开(公告)号：EP0623914A1

    公开(公告)日：1994-11-09

    申请号：EP94106987.4

    申请日：1994-05-04

    申请人： CSELT Centro Studi e Laboratori Telecomunicazioni S.p.A.

    发明人： Albesano, Dario ,  Gemello, Roberto ,  Mana, Franco

    IPC分类号： G10L5/06 ,  G10L7/08 ,  G10L9/06

    CPC分类号： G10L15/16

    摘要： The method for speaker independent isolated word recognition is based on a hybrid recognition system, which uses neural networks, availing itself of its parallel processing to improve recognition and optimize system for what concerns time and memory while it keeps some of the consolidated aspects of recognition techniques.
    Complete words are modeled with left-to-right Markov model automata with recursion on states, each of which corresponds to an acoustic portion of the word, and recognition is obtained by performing a dynamic programming according to the Viterbi algorithm on all automata in order to detect the one having the minimum cost path to which corresponds the recognized word,
    the emission probabilities being computed through a neural network with feedback, trained in an original way, and the transition probabilities being suitably estimated.

    摘要翻译： 扬声器孤立词识别的方法是基于一种混合识别系统，它采用神经网络，出于自身需要的并行处理来提高识别和优化所关注的时间和内存，同时它保持一定的识别技术的整合方面系统 ， 完整的单词建模与左到右马尔可夫模型自动机与美国，每一个在字的声部对应于递归，识别是通过以便对所有自动执行动态编程雅鼎的维特比算法来获得 检测具有最小成本路径而对应识别的单词，正在通过具有反馈的神经网络计算出的发射概率，在原始训练方式中，和转换概率被适当地估计出的一个。