Indonesian Automatic Speech Recognition system using CMUSphinx toolkit and limited dataset
Hamdan Prakoso, Ridi Ferdiana, Rudy Hartanto · 2016
Building Automatic Speech Recognition (ASR) needs acoustic model, language model and dictionary for intended language, which is also applied for Indonesian ASR. In this paper, Indonesian ASR was built using CMUSphinx toolkit (a Hidden Markov Model based ASR tool) with limited dataset. We use digit corpus and own made language model to trained with the limited dataset. We also investigated the implementation of trained acoustic model by examine it in different SNR condition to several people. The best achievement of word error accuracy of the acoustic model is 86% on average. By examine it in different SNR condition, we got maximum accuracy of 80% on 27.764 dB environment.