Improvements to the pruning behavior of DNN acoustic models
نویسنده
چکیده
This paper examines two strategies that improve the beam pruning behavior of DNN acoustic models with only a negligible increase in model complexity. By augmenting the boosted MMI loss function used in sequence training with the weighted cross-entropy error, we achieve a real time factor (RTF) reduction of more than 13%. By directly incorporating a transition model into the DNN, which leads to a parameter size increase of less than 0.017%, we achieve a RTF reduction of 16%. Combining both techniques results in a RTF reduction of more than 23%. Both strategies, and their combination, also lead to small but statistically significant word error rate reductions.
منابع مشابه
Investigation on acoustic behavior of acoustic porous absorbers to absorb sound energy and transmission loss index
In this study, the acoustic properties of porous absorbents with different porosity levels have been evaluated using different mathematical models. These models use one or more parameters of materials for calculating acoustic characteristics. In all of these models, materials are considered as equivalent fluid and reactionary characteristics have not been taken into account.
متن کاملImproving Phoneme Sequence Recognition using Phoneme Duration Information in DNN-HSMM
Improving phoneme recognition has attracted the attention of many researchers due to its applications in various fields of speech processing. Recent research achievements show that using deep neural network (DNN) in speech recognition systems significantly improves the performance of these systems. There are two phases in DNN-based phoneme recognition systems including training and testing. Mos...
متن کاملRestructuring of deep neural network acoustic models with singular value decomposition
Recently proposed deep neural network (DNN) obtains significant accuracy improvements in many large vocabulary continuous speech recognition (LVCSR) tasks. However, DNN requires much more parameters than traditional systems, which brings huge cost during online evaluation, and also limits the application of DNN in a lot of scenarios. In this paper we present our new effort on DNN aiming at redu...
متن کاملRobust speech recognition using DNN-HMM acoustic model combining noise-aware training with spectral subtraction
Recently, acoustic models based on deep neural notworks (DNNs) have been introduced and showed dramatic improvements over acoustic models based on GMM in a variety of tasks. In this paper, we considered the improvement of noise robustness of DNN. Inspired by Missing Feature Theory and static noise aware training, we proposed an approach that uses a noise-suppressed acoustic feature and estimate...
متن کاملEnsemble of Jointly Trained Deep Neural Network-Based Acoustic Models for Reverberant Speech Recognition
Distant speech recognition is a challenge, particularly due to the corruption of speech signals by reverberation caused by large distances between the speaker and microphone. In order to cope with a wide range of reverberations in real-world situations, we present novel approaches for acoustic modeling including an ensemble of deep neural networks (DNNs) and an ensemble of jointly trained DNNs....
متن کامل