An end-to-end Spanish lipreading system based on the CTC/Attention architecture achieves state-of-the-art WER of 24.8% on VLRF and 34.5%/59.5% on LIP-RTVE speaker-dependent/independent partitions, with an ablation study and a new four-corpus benchmark.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Evaluation of End-to-End Continuous Spanish Lipreading in Different Data Conditions
An end-to-end Spanish lipreading system based on the CTC/Attention architecture achieves state-of-the-art WER of 24.8% on VLRF and 34.5%/59.5% on LIP-RTVE speaker-dependent/independent partitions, with an ablation study and a new four-corpus benchmark.