Search papers, labs, and topics across Lattice.
Carnegie Mellon University
1
0
3
22
Achieve single-pass alignment of multi-talker speech – a feat previously impossible – by modeling overlaps as shuffles.
