MaSS (Multilingual corpus of Sentence-aligned Spoken utterances) is an extension of the CMU Wilderness Multilingual Speech Dataset, a speech dataset based on recorded readings of the New Testament.
15 PAPERS • 3 BENCHMARKS