The MALACH corpus is packaged with benchmark train/test partitions and a 21.7% WER baseline using LSTM acoustic and language models.
To obtain better complementar- ity, a 64-dimensional filter bank was created and used to extract logmel, delta, and delta-delta parameters
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Challenging the Boundaries of Speech Recognition: The MALACH Corpus
The MALACH corpus is packaged with benchmark train/test partitions and a 21.7% WER baseline using LSTM acoustic and language models.