Frozen speech recognition and machine translation models can be aligned by a small connector network to perform end-to-end speech translation, and the connector also serves as a domain adapter.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Aligning Pre-trained Models for Spoken Language Translation
Frozen speech recognition and machine translation models can be aligned by a small connector network to perform end-to-end speech translation, and the connector also serves as a domain adapter.