MLS is a new large-scale multilingual speech corpus derived from LibriVox with 44.5k hours of English and 6k hours across seven other languages, plus baseline ASR and LM models.
Identification of common molecular subsequences,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2020 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
MLS: A Large-Scale Multilingual Dataset for Speech Research
MLS is a new large-scale multilingual speech corpus derived from LibriVox with 44.5k hours of English and 6k hours across seven other languages, plus baseline ASR and LM models.