AppTek's Submission to the IWSLT 2022 Isometric Spoken Language Translation Task. (arXiv:2205.05807v1 [cs.CL])

To participate in the Isometric Spoken Language Translation Task of the IWSLT
2022 evaluation, constrained condition, AppTek developed neural
Transformer-based systems for English-to-German with various mechanisms of
length control, ranging from source-side and target-side pseudo-tokens to
encoding of remaining length in characters that replaces positional encoding.
We further increased translation length compliance by sentence-level selection
of length-compliant hypotheses from different system variants, as well as
rescoring of N-best candidates from a single system. Length-compliant
back-translated and forward-translated synthetic data, as …

arxiv language language translation translation