Bayesian learning of LF-MMI trained time delay neural networks for speech recognition
Discriminative training techniques define state-of-the-art performance for automatic speech
recognition systems. However, they are inherently prone to overfitting, leading to poor …
recognition systems. However, they are inherently prone to overfitting, leading to poor …
Conditional random fields for integrating local discriminative classifiers
J Morris, E Fosler-Lussier - IEEE Transactions on Audio …, 2008 - ieeexplore.ieee.org
Conditional random fields (CRFs) are a statistical framework that has recently gained in
popularity in both the automatic speech recognition (ASR) and natural language processing …
popularity in both the automatic speech recognition (ASR) and natural language processing …
Neural networks for distant speech recognition
S Renals, P Swietojanski - 2014 4th joint workshop on hands …, 2014 - ieeexplore.ieee.org
Distant conversational speech recognition is challenging owing to the presence of multiple,
overlapping talkers, additional non-speech acoustic sources, and the effects of …
overlapping talkers, additional non-speech acoustic sources, and the effects of …
[PDF][PDF] On the use of morphological analysis for dialectal Arabic speech recognition.
Arabic has a large number of affixes that can modify a stem to form words. In automatic
speech recognition (ASR) this leads to a high out-of-vocabulary (OOV) rate for typical …
speech recognition (ASR) this leads to a high out-of-vocabulary (OOV) rate for typical …
Multilingual MRASTA features for low-resource keyword search and speech recognition systems
Z Tüske, D Nolden, R Schlüter… - 2014 IEEE International …, 2014 - ieeexplore.ieee.org
This paper investigates the application of hierarchical MRASTA bottleneck (BN) features for
under-resourced languages within the IARPA Babel project. Through multilingual training of …
under-resourced languages within the IARPA Babel project. Through multilingual training of …
Random rough subspace based neural network ensemble for insurance fraud detection
W Xu, S Wang, D Zhang, B Yang - 2011 Fourth International …, 2011 - ieeexplore.ieee.org
In this paper, a random rough subspace based neural network ensemble method is
proposed for insurance fraud detection. In this method, rough set reduction is firstly …
proposed for insurance fraud detection. In this method, rough set reduction is firstly …
[PDF][PDF] Discriminative methods for noise robust speech recognition: A CHiME challenge benchmark
The recently introduced second CHiME challenge is a difficult two-microphone speech
recognition task with non-stationary interference. Current approaches in the source …
recognition task with non-stationary interference. Current approaches in the source …
Discriminative training using noise robust integrated features and refined HMM modeling
The classical approach to build an automatic speech recognition (ASR) system uses
different feature extraction methods at the front end and various parameter classification …
different feature extraction methods at the front end and various parameter classification …
[PDF][PDF] Investigations on error minimizing training criteria for discriminative training in automatic speech recognition.
Discriminative training criteria have been shown to consistently outperform maximum
likelihood trained speech recognition systems. In this paper we employ the Minimum …
likelihood trained speech recognition systems. In this paper we employ the Minimum …
IBM MASTOR: Multilingual automatic speech-to-speech translator
Y Gao, B Zhou, L Gu, R Sarikaya… - … on Acoustics Speech …, 2006 - ieeexplore.ieee.org
In this paper, we describe the IBM MASTOR systems which handle spontaneous free-form
speech-to-speech translation on both laptop and hand-held PDAs. Challenges include …
speech-to-speech translation on both laptop and hand-held PDAs. Challenges include …