IMaSC: A Malayalam Speech Corpus for High-Quality Text-to-Speech Synthesis

Abstract

Modern text-to-speech (TTS) systems use deep learning to synthesize speech increasingly approaching human quality, but they require a database of high-quality audio-text sentence pairs for training. Malayalam, the official language of the Indian state of Kerala and spoken by 35+ million people, is a low-resource language in terms of available corpora for TTS systems. In this paper, we present IMaSC, a Malayalam text and speech corpora containing 49 hours and 37 minutes of recorded speech. With 8 speakers and a total of 34,473 text-audio pairs, IMaSC is larger than every other publicly available alternative. We evaluated the database by using it to train TTS models for each speaker based on a modern deep learning architecture. With an average mean opinion score of 4.50, we find that the synthesized speech of our model is close to human quality.

Resources

Details