Open-source Multi-speaker Speech Corpora for Building Gujarati, Kannada, Malayalam, Marathi, Tamil and Telugu Speech Synthesis Systems

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 6494–6503 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 6494 Open-source Multi-speaker Speech Corpora for Building Gujarati, Kannada, Malayalam, Marathi, Tamil and Telugu Speech Synthesis Systems Fei He, Shan-Hui Cathy Chu, Oddur Kjartansson, Clara Rivera, Anna Katanova, Alexander Gutkin, Is¸ın Demirs¸ahin, Cibu Johny, Martin Jansche†, Supheakmungkol Sarin, Knot Pipatsrisawat Google Research Singapore, United States and United Kingdom {oddur,rivera,agutkin,isin,cibu,mungkol,thammaknot}@google.com Abstract We present free high quality multi-speaker speech corpora for Gujarati, Kannada, Malayalam, Marathi, Tamil and Telugu, which are six of the twenty two official languages of India spoken by 374 million native speakers. The datasets are primarily intended for use in text-to-speech (TTS) applications, such as constructing multilingual voices or being used for speaker or language adaptation. Most of the corpora (apart from Marathi, which is a female-only database) consist of at least 2,000 recorded lines from female and male native speakers of the language. We present the methodological details behind corpora acquisition, which can be scaled to acquiring data for other languages of interest. We describe the experiments in building a multilingual text-to-speech model that is constructed by combining our corpora. Our results indicate that using these corpora results in good quality voices, with Mean Opinion Scores (MOS) > 3.6, for all the languages tested. We believe that these resources, released with an open-source license, and the described methodology will help in the progress of speech applications for the languages described and aid corpora development for other, smaller, languages of India and beyond. Keywords: speech corpora, low-resource, text-to-speech, Gujarati, Kannada, Marathi, Malayalam, Tamil, Telugu, open-source 1. Introduction Voice communication is one of the most natural and con- venient modes of human interaction. As technologies in this field have advanced, computer applications that can use natural speech to communicate with users have become in- creasingly popular. In this work, we deal with six out of the twenty two official languages of India (Mohanty, 2006; Mohanty, 2010): Gujarati, Kannada, Malayalam, Marathi, Tamil and Telugu, which have combined speaker popula- tion of close to 400 million people. Although the situa- tion with the speech corpora availability for these languages has been improving, these languages are still considered by many to be low-resource (Besacier et al., 2014; Srivastava et al., 2018). Furthermore, the resources available for build- ing speech technology (and text-to-speech (TTS) applica- tions, in particular) for these languages are still relatively scarce compared to those of Hindi, the most widely-spoken language of India. We had published the Bangla speech corpora previously (Gutkin et al., 2016; Kjartansson et al., 2018) and these are the next six largest languages of India. There are four main resource components required to con- struct a classical TTS system: a speech corpus, a phonolog- ical inventory, a pronunciation lexicon and a text normaliza- tion front-end. Among these four components, speech cor- pora are usually the most expensive to develop. In the con- ventional approach, one would need to carefully design …

Content truncated to 500 words