Joint RNN transducer for ASR and speaker diarization reduces word-level diarization error rate from 15.8% to 2.2% on medical conversations.
Problem Formulation and Proposed Solution Many machine learning tasks can be expressed as mapping an input sequence into an output sequence
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Joint Speech Recognition and Speaker Diarization via Sequence Transduction
Joint RNN transducer for ASR and speaker diarization reduces word-level diarization error rate from 15.8% to 2.2% on medical conversations.