Reproduce “Indonesian-English Code-Switching Speech Synthesizer Utilizing Multilingual STEN-TTS and Bert LID” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ahmad Alfani Handoyo, Chung Tran, Dessi Puji Lestari, Sakriani Sakti
The authors present CS-TTS, a system for Indonesian–English code-switching speech synthesis. It identifies each word’s language, applies the corresponding pronunciation rules, and removes STEN-TTS’s sentence-level language embedding. On the seven-case TestSet, it reports total MOS of 3.379, versus 2.163 and 1.841 for English and Indonesian baselines. Its WER is 17.43% versus 36.44% and 42.18%, in a small listening study.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.