隐马尔可夫过程(Hidden Markov Model, HMM)是一种统计模型,用于描述一个系统在不同状态间转换且这些状态不可直接观测的情况。该模型通过观察序列推断隐藏的状态序列,在语音识别、自然语言处理等领域有广泛应用。
Hidden Markov processes (HMPs) were introduced into the statistics literature as early as 1966. Starting in the mid-1970s, HMPs have been utilized in speech recognition, which is likely the earliest application of these models outside a purely mathematical context.