Picture for P. C. Woodland

P. C. Woodland

Transformer Language Models with LSTM-based Cross-utterance Information Representation

Add code
Feb 12, 2021
Figure 1 for Transformer Language Models with LSTM-based Cross-utterance Information Representation
Figure 2 for Transformer Language Models with LSTM-based Cross-utterance Information Representation
Figure 3 for Transformer Language Models with LSTM-based Cross-utterance Information Representation
Figure 4 for Transformer Language Models with LSTM-based Cross-utterance Information Representation
Viaarxiv icon

Content-Aware Speaker Embeddings for Speaker Diarisation

Add code
Feb 12, 2021
Figure 1 for Content-Aware Speaker Embeddings for Speaker Diarisation
Figure 2 for Content-Aware Speaker Embeddings for Speaker Diarisation
Figure 3 for Content-Aware Speaker Embeddings for Speaker Diarisation
Figure 4 for Content-Aware Speaker Embeddings for Speaker Diarisation
Viaarxiv icon

Cross-Utterance Language Models with Acoustic Error Sampling

Add code
Aug 19, 2020
Figure 1 for Cross-Utterance Language Models with Acoustic Error Sampling
Figure 2 for Cross-Utterance Language Models with Acoustic Error Sampling
Figure 3 for Cross-Utterance Language Models with Acoustic Error Sampling
Figure 4 for Cross-Utterance Language Models with Acoustic Error Sampling
Viaarxiv icon

Combining Natural Gradient with Hessian Free Methods for Sequence Training

Add code
Oct 03, 2018
Figure 1 for Combining Natural Gradient with Hessian Free Methods for Sequence Training
Figure 2 for Combining Natural Gradient with Hessian Free Methods for Sequence Training
Figure 3 for Combining Natural Gradient with Hessian Free Methods for Sequence Training
Figure 4 for Combining Natural Gradient with Hessian Free Methods for Sequence Training
Viaarxiv icon