Statistical Language Model Adaptation for Persian Speech Recognition

Language models are important in various applications especially in speech recognition.Extracting n-gram statistics is a prevalent approach for statistical language modeling. But Traditional n-gram language models suffer from insufficient long-distance information and have crucial dependency on the training corpus. The aim of language model adaptation is to exploit specific, albeit limited, knowledge about the recognition task to compensate for this mismatch. This paper presentsan overview of the major adaptation approaches proposed to deal with this issue and we implement these approaches for Persian continuous speech recognition

کلیدواژه ها:

Speech Recognition ، Statistical Language Model Adaptation ، Corpus