Backpropagation training for multilayer conditional random field based phone recognition
- 1 January 2010
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5534-5537
Abstract
Conditional random fields (CRFs) have recently found increased popularity in automatic speech recognition (ASR) applications. CRFs have previously been shown to be effective combiners of posterior estimates from multilayer perceptrons (MLPs) in phone and word recognition tasks. In this paper, we describe a novel hybrid Multilayer-CRF structure (ML-CRF), where a MLP-like hidden layer serves as input to the CRF; moreover, we propose a technique for directly training the ML-CRF to optimize a conditional log-likelihood based criterion, based on error backpropagation. The proposed technique thus allows for the implicit learning of suitable feature functions for the CRF. We present results for initial phone recognition experiments on the TIMIT database that indicate that our proposed method is a promising approach for training CRFs.Keywords
This publication has 6 references indexed in Scilit:
- Discriminative Input Stream Combination for Conditional Random Field Phone RecognitionIEEE Transactions on Audio, Speech, and Language Processing, 2009
- Lattice-based optimization of sequence classification criteria for neural-network acoustic modeling2009
- Speech Recognition Using Augmented Conditional Random FieldsIEEE Transactions on Audio, Speech, and Language Processing, 2009
- Conditional Random Fields for Integrating Local Discriminative ClassifiersIEEE Transactions on Audio, Speech, and Language Processing, 2008
- Shallow parsing with conditional random fields2003
- Speaker-independent phone recognition using hidden Markov modelsIEEE Transactions on Acoustics, Speech, and Signal Processing, 1989