Kannada Continuous Speech Recognition Using Deep Learning

Automatic Speech Recognition (ASR) is a prevalent approach for attaining human-machine interaction by enabling machines to transcribe speech data. We propose a Continuous Speech Recognition model in the Kannada language using deep learning techniques such as Convolutional Neural Networks (CNN) and Bidirectional Gated Recurrent Units (Bi-GRU). The model was trained and validated using 100 and 20 h of data, respectively. The experiment has generated encouraging results with a Character Error Rate (CER) of 15.62% and a Word Error Rate of 34.47% (WER).