A Small Vocabulary Speech Recognition for Gujarati
Himanshu Patel, Paresh Virparia · International Journal of Advanced Research in Computer Science · 2011
In this paper we describe how to recognize small vocabulary speech in Guajarati language using English Speech Recognition Engine. We present a technique for fast bootstrapping of initial phone models of a Gujarati language. The training data for the Gujarati language is aligned using an existing speech recognition engine for English language. This aligned data is used to obtain the initial acoustic models for the phones of the Gujarati language. Following this approach requires less training data. As is inherent in phonetic languages, rules generally capture the mapping of spelling to phonemes very well. This paper described technique used and result obtained. Keywords: Gujarati Speech Recognition, Speech Recognition Engine, Acoustic model, language model.