Discriminative Phonetic Recognition with Conditional Random Fields
نویسنده
چکیده
A Conditional Random Field is a mathematical model for sequences that is similar in many ways to a Hidden Markov Model, but is discriminative rather than generative in nature. In this paper, we explore the application of the CRF model to ASR processing of discriminative phonetic features by building a system that performs first-pass phonetic recognition using discriminatively trained phonetic features. With this system, we show that this CRF model trained on only monophone labels achieves an accuracy level in a phone recognition task that is close to that of an HMM model that has been trained on triphone labels.
منابع مشابه
Combining phonetic attributes using conditional random fields
A Conditional Random Field is a mathematical model for sequences that is similar in many ways to a Hidden Markov Model, but is discriminative rather than generative in nature. Here we explore the application of the CRF model to ASR processing by building a system that performs first-pass phonetic recogintion using discriminatively trained phonetic attributes. This system achieves an accuracy le...
متن کاملCombining phonetic attributes usin
A Conditional Random Field is a mathematical model for sequences that is similar in many ways to a Hidden Markov Model, but is discriminative rather than generative in nature. In this paper, we explore the application of the CRF model to ASR processing of discriminative phonetic features by building a system that performs first-pass phonetic recognition using discriminatively trained phonetic f...
متن کاملDeep-structured hidden conditional random fields for phonetic recognition
We extend our earlier work on deep-structured conditional random field (DCRF) and develop deep-structured hidden conditional random field (DHCRF). We investigate the use of this new sequential deep-learning model for phonetic recognition. DHCRF is a hierarchical model in which the final layer is a hidden conditional random field (HCRF) and the intermediate layers are zero-th-order conditional r...
متن کاملSegmental conditional random fields with deep neural networks as acoustic models for first-pass word recognition
Discriminative segmental models, such as segmental conditional random fields (SCRFs), have been successfully applied to speech recognition recently in lattice rescoring to integrate detectors across different levels of units, such as phones and words. However, the lattice generation has been constrained by a baseline decoder, typically a frame-based hybrid HMMDNN system, which still suffers fro...
متن کاملImproved Discriminative Model for View- Invariant Human Action Recognition
Recognizing human actions play an important role in applications like video surveillance. The recent past has witnessed an increasing research on view-invariant action recognition. Huang et al. proposed a framework based on discriminative model for human action recognition. This model uses STIP (Space – Time Interest Point) to extract motion features and view invariants. Then a discriminative m...
متن کامل