Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Thomas Hain

Multiple-hypothesis CTC-based semi-supervised adaptation of end-to-end speech recognition


Mar 31, 2021
Cong-Thanh Do, Rama Doddipatla, Thomas Hain

* Accepted at ICASSP 2021 

  Access Paper or Ask Questions

T-vectors: Weakly Supervised Speaker Identification Using Hierarchical Transformer Model


Oct 29, 2020
Yanpei Shi, Mingjie Chen, Qiang Huang, Thomas Hain

* Submitted to ICASSP2021. arXiv admin note: text overlap with arXiv:2005.07817 

  Access Paper or Ask Questions

Exploration of Audio Quality Assessment and Anomaly Localisation Using Attention Models


May 16, 2020
Qiang Huang, Thomas Hain

* Submitted to InterSpeech 2020 

  Access Paper or Ask Questions

Speaker Re-identification with Speaker Dependent Speech Enhancement


May 15, 2020
Yanpei Shi, Qiang Huang, Thomas Hain

* Submitted to Interspeech2020 

  Access Paper or Ask Questions

Weakly Supervised Training of Hierarchical Attention Networks for Speaker Identification


May 15, 2020
Yanpei Shi, Qiang Huang, Thomas Hain

* Submitted to Interspeech2020 

  Access Paper or Ask Questions

Supervised Speaker Embedding De-Mixing in Two-Speaker Environment


Jan 14, 2020
Yanpei Shi, Thomas Hain

* Submitted to Odyssey 2020 

  Access Paper or Ask Questions

Robust Speaker Recognition Using Speech Enhancement And Attention Model


Jan 14, 2020
Yanpei Shi, Qiang Huang, Thomas Hain

* Submitted to Odyssey 2020 

  Access Paper or Ask Questions

H-VECTORS: Utterance-level Speaker Embedding Using A Hierarchical Attention Model


Oct 19, 2019
Yanpei Shi, Qiang Huang, Thomas Hain


  Access Paper or Ask Questions

Contextual Joint Factor Acoustic Embeddings


Oct 16, 2019
Yanpei Shi, Qiang Huang, Thomas Hain


  Access Paper or Ask Questions

Improving Robustness In Speaker Identification Using A Two-Stage Attention Model


Sep 24, 2019
Yanpei Shi, Qiang Huang, Thomas Hain


  Access Paper or Ask Questions

Latent Dirichlet Allocation Based Acoustic Data Selection for Automatic Speech Recognition


Jul 02, 2019
Mortaza, Doulaty, Thomas Hain

* Proc. of Interspeech (2019), Graz, Austria 

  Access Paper or Ask Questions

Automatic Genre and Show Identification of Broadcast Media


Jun 10, 2016
Mortaza Doulaty, Oscar Saz, Raymond W. M. Ng, Thomas Hain

* Proc. of 17th Interspeech (2016), San Francisco, California, USA 

  Access Paper or Ask Questions

The 2015 Sheffield System for Transcription of Multi-Genre Broadcast Media


Dec 21, 2015
Oscar Saz, Mortaza Doulaty, Salil Deena, Rosanna Milner, Raymond W. M. Ng, Madina Hasan, Yulan Liu, Thomas Hain

* IEEE Automatic Speech Recognition and Understanding Workshop (ASRU 2015), 13-17 Dec 2015, Scottsdale, Arizona, USA 

  Access Paper or Ask Questions

Latent Dirichlet Allocation Based Organisation of Broadcast Media Archives for Deep Neural Network Adaptation


Nov 16, 2015
Mortaza Doulaty, Oscar Saz, Raymond W. M. Ng, Thomas Hain

* IEEE Automatic Speech Recognition and Understanding Workshop (ASRU 2015), 13-17 Dec 2015, Scottsdale, Arizona, USA 

  Access Paper or Ask Questions

The USFD Spoken Language Translation System for IWSLT 2014


Sep 13, 2015
Raymond W. M. Ng, Mortaza Doulaty, Rama Doddipatla, Wilker Aziz, Kashif Shah, Oscar Saz, Madina Hasan, Ghada AlHarbi, Lucia Specia, Thomas Hain

* Proc. of 11th International Workshop on Spoken Language Translation (SLT 2014) 86-91, Lake Tahoe, USA, December 4th and 5th, 2014 

  Access Paper or Ask Questions

Unsupervised Domain Discovery using Latent Dirichlet Allocation for Acoustic Modelling in Speech Recognition


Sep 08, 2015
Mortaza Doulaty, Oscar Saz, Thomas Hain

* 16th Interspeech.Proc. (2015) 3640-3644, Dresden, Germany 

  Access Paper or Ask Questions

Data-selective Transfer Learning for Multi-Domain Speech Recognition


Sep 08, 2015
Mortaza Doulaty, Oscar Saz, Thomas Hain

* 16th Interspeech.Proc. (2015) 2897-2901 

  Access Paper or Ask Questions