The MERL/MELCO/TUM system for the REVERB Challenge using Deep Recurrent Neural Network Feature Enhancement
Proceedings REVERB Workshop, held in conjunction with ICASSP 2014 and HSCMA 2014, Florence, Italy
second best result
2014
Feature Enhancement by Deep LSTM Networks for ASR in Reverberant Multisource Environments
Computer Speech and Language, Elsevier
2014
Vol. 28
Jul
pp. 888–902
On-line NMF-based Stereo Up-Mixing of Speech Improves Perceived Reduction of Non-Stationary Noise
Proceedings AES 53rd International Conference Semantic Audio (K. Brandenburg and M. Sandler, eds.), Lonon, UK
Best Student Paper Award
2014
Robust Speech Recognition using Long Short-Term Memory Recurrent Neural Networks for Hybrid Acoustic Modelling
Proc. INTERSPEECH 2014, Singapore, Singapore
2014
On the Influence of Alcohol Intoxication on Speaker Recognition
Proc. AES 53rd International Conference Semantic Audio, AES, London, UK
2014
Memory-Enhanced Neural Networks and NMF for Robust ASR
Audio, Speech and Language Processing, IEEE/ACM Transactions on
2014
22
6
Jun
1037 - 1046
The TUM system for the REVERB Challenge: Recognition of Reverberated Speech using Multi-Channel Correlation Shaping Dereverberation and BLSTM Recurrent Neural Networks
Proceedings REVERB Workshop, held in conjunction with ICASSP 2014 and HSCMA 2014, Florence, Italy
2014
Acoustic Gaitbased Person Identification using Hidden Markov Models
Proceedings of the Personality Mapping Challenge & Workshop (MAPTRAITS 2014), Satellite of the 16th ACM International Conference on Multimodal Interaction (ICMI 2014), Istanbul, Turkey
2014
Investigating NMF Speech Enhancement for Neural Network based Acoustic Models
Proc. INTERSPEECH 2014, ISCA, Singapore, Singapore
2014