Picture for Thomas Zenkel

Thomas Zenkel

End-to-End Neural Word Alignment Outperforms GIZA++

Add code
Apr 30, 2020
Figure 1 for End-to-End Neural Word Alignment Outperforms GIZA++
Figure 2 for End-to-End Neural Word Alignment Outperforms GIZA++
Figure 3 for End-to-End Neural Word Alignment Outperforms GIZA++
Figure 4 for End-to-End Neural Word Alignment Outperforms GIZA++
Viaarxiv icon

Adding Interpretable Attention to Neural Translation Models Improves Word Alignment

Add code
Jan 31, 2019
Figure 1 for Adding Interpretable Attention to Neural Translation Models Improves Word Alignment
Figure 2 for Adding Interpretable Attention to Neural Translation Models Improves Word Alignment
Figure 3 for Adding Interpretable Attention to Neural Translation Models Improves Word Alignment
Figure 4 for Adding Interpretable Attention to Neural Translation Models Improves Word Alignment
Viaarxiv icon

Subword and Crossword Units for CTC Acoustic Models

Add code
Jun 18, 2018
Figure 1 for Subword and Crossword Units for CTC Acoustic Models
Figure 2 for Subword and Crossword Units for CTC Acoustic Models
Figure 3 for Subword and Crossword Units for CTC Acoustic Models
Figure 4 for Subword and Crossword Units for CTC Acoustic Models
Viaarxiv icon

Comparison of Decoding Strategies for CTC Acoustic Models

Add code
Aug 15, 2017
Figure 1 for Comparison of Decoding Strategies for CTC Acoustic Models
Figure 2 for Comparison of Decoding Strategies for CTC Acoustic Models
Figure 3 for Comparison of Decoding Strategies for CTC Acoustic Models
Viaarxiv icon