Semi-supervised Learning with Sparse Autoencoders in Phone Classification
We propose the application of a semi-supervised learning method to improve the performance of acoustic modelling for automatic speech recognition based on deep neural net- works. As opposed to unsupervised initialisation followed by supervised fine tuning, our method takes advantage of both unlabelled and labelled data simultaneously through mini- batch stochastic gradient descent. We tested the method with varying proportions of labelled vs unlabelled observations in frame-based phoneme classification on the TIMIT database. Our experiments show that the method outperforms standard supervised training for an equal amount of labelled data and provides competitive error rates compared to state-of-the-art graph-based semi-supervised learning techniques.
Code (0)
등록된 구현이 없습니다.
Tasks
Acoustic ModellingAutomatic Speech RecognitionAutomatic Speech Recognition (ASR)ClassificationGeneral Classificationspeech-recognitionSpeech RecognitionSimilar Papers 제목 키워드 기반
Augmenting Variational Autoencoders with Sparse Labels: A Unified Framework for Unsupervised, Semi-(un)supervised, and Supervised Learning
We present a new flavor of Variational Autoencoder (VAE) that interpolates seamlessly between unsupervised, semi-supervised and fully supervised learning domains. We show that unlabeled datapoints not only boost unsuperv…
ClassificationDecoderGeneral ClassificationLearning Consistent Deep Generative Models from Sparse Data via Prediction Constraints
We develop a new framework for learning variational autoencoders and other deep generative models that balances generative and discriminative goals. Our framework optimizes model parameters to maximize a variational lowe…
General Classificationimage-classificationImage ClassificationvONTSS: vMF based semi-supervised neural topic modeling with optimal transport
Recently, Neural Topic Models (NTM), inspired by variational autoencoders, have attracted a lot of research interest; however, these methods have limited applications in the real world due to the challenge of incorporati…
ClassificationDiversitytext-classificationText Classification+2Semi-supervised source localization in reverberant environments with deep generative modeling
We propose a semi-supervised approach to acoustic source localization in reverberant environments based on deep generative modeling. Localization in reverberant environments remains an open challenge. Even with large dat…
Learning Consistent Deep Generative Models from Sparsely Labeled Data
We consider training deep generative models toward two simultaneous goals: discriminative classification and generative modeling using an explicit likelihood. While variational autoencoders (VAEs) offer a promising solut…
image-classificationImage Classification