The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In this paper, we present a novel multi-photo-based framework to solve a self-portrait enhancement problem we call “1+2 problem”, in which a self-portrait photo is enhanced with the help of two multiple photos that share the same scene and similar shooting time. The key idea is to exploit the extra information of these two photos to overcome the limited field of view and poor illumination of the target...
This paper proposes an approach to perform accent adaptation by using accent dependent bottleneck (BN) features to improve the performance of multi-accent Mandarin speech recognition system. The architecture of the adaptation uses two neural networks. First, deep neural network (DNN) acoustic model acts as a feature extractor which is used to extract accent dependent BN (BN-DNN) features. The input...
In fingerprint recognition system, minutiae-based matching algorithms are most intensively researched. However, in most minutia-based methods, the similarity score is given based on the main score of matched minutiae. And the boosted information is not effectively used in the final similarity score computation. Based on the observation, we extract several features as the supplementary scores. And...
This paper proposes a cascading deep neural network (DNN) structure for speech synthesis system that consists of text-to-bottleneck (TTB) and bottleneck-to-speech (BTS) models. Unlike conventional single structure that requires a large database to find complicated mapping rules between linguistic and acoustic features, the proposed structure is very effective even if the available training database...
Recently, a deep beamforming (BF) network was proposed to predict BF weights from phase-carrying features, such as generalized cross correlation (GCC). The BF network is trained jointly with the acoustic model to minimize automatic speech recognition (ASR) cost function. In this paper, we propose to replace GCC with features derived from input signals' spatial covariance matrices (SCM), which contain...
In this paper, we propose a framework that fuses textual and visual features of user generated social media data to mine the distribution of user interests. The proposed framework consists of three steps: feature extraction, model training, and user interest mining. We choose boards from popular users on Pinterest to collect training and test data. For each pin we extract the term-document matrices...
We propose a voice conversion framework to map the speech features of a source speaker to a target speaker based on deep neural networks (DNNs). Due to a limited availability of the parallel data needed for a pair of source and target speakers, speech synthesis and dynamic time warping are utilized to construct a large parallel corpus for DNN training. With a small corpus to train DNNs, a lower log...
We propose a sudden-noise suppression method for speech recognition using a phase linearity feature for noise detection. Our investigation of sound data recorded in actual retail stores shows that short, sudden noises are dominant in such environments. We also confirm the negative effect of such noises on speech recognition performance. Our method addresses this problem by focusing on sudden noises...
In this study, we investigate on the learning behaviors of DNN by explicit feature transformations. As a demonstration, linear and logarithm transformations, corresponding to the amplitude spectra and log-power spectra, are compared with the same minimum mean squared error (MMSE) objective function for optimizing DNN parameters. Based on the experimental analysis of the DNN learning behaviors, we...
In this paper, we propose a semi-global matching method based on image segmentation. We perform a k-means clustering algorithm in only left image as image segmentation. Then, to improve result of image segmentation, we integrate adjacent and small labels along edges of objects. After that, we extract feature points to estimate the disparity range in each label, and add weights to the disparity range...
In this paper, we develop an algorithm for depth image super-resolution from RGB-D images, which are acquired under different imaging conditions so that we can combine them to improve the image quality with precise 3D registration. We focus on how to increase the resolution and quality of depth images by combining multiple RGB-D images and using the deep learning technique. In the proposed solution,...
Recently, the deep neural networks (DNNs) are successfully adopted into the voice activity detection (VAD) area. However, the performance of the DNN-based VAD is still unsatisfactory in noise environments where the feature subspace of the training database and the test environments are not matched with each other. In this paper, we propose a local feature shift technique which normalizes the feature...
In this paper, a novel Dynamic Convolutional Neural Network (D-CNN) is proposed using sensor data for activity recognition. Sensor data collected for activity recognition is usually not well-aligned. It may also contains noises and variations from different persons. To overcome these challenges, Gaussian Mixture Models (GMM) is exploited to capture the distribution of each activity. Then, sensor data...
Using speech or text to predict articulatory movements can have potential benefits for speech related applications. Many approaches have been proposed to solve the acoustic-to-articulatory inversion problem, which is much more than the exploration for predicting articulatory movements from text. In this paper, we investigate the feasibility of using deep neural network (DNN) for articulartory movement...
Given the increasing attention paid to speech emotion classification in recent years, this work presents a novel speech emotion classification approach based on the multiple kernel Gaussian process. Two major aspects of a classification problem that play an important role in classification accuracy are addressed, i.e. feature extraction and classification. Prosodic features and other features widely...
Recently, deep and/or recurrent neural networks (DNNs/RNNs) have been employed for voice conversion, and have significantly improved the performance of converted speech. However, DNNs/RNNs generally require a large amount of parallel training data (e.g., hundreds of utterances) from source and target speakers. It is expensive to collect such a large amount of data, and impossible in some applications,...
We propose to detect mispronunciations in a language learners speech via a discriminatively trained DNN in the phonetic space. The posterior probabilities of “senones” populated in a decision tree are trained and predicted speaker independently. Acoustic features of each input segment (with preceding and succeeding contexts of several frames) are mapped unto the whole set of senones in their corresponding...
Speaker verification suffers from serious performance degradation under speaking rate mismatch condition. This degradation can be largely attributed to the spectrum distortion caused by different speaking rates. This paper proposes a feature transform approach which projects speech features in slow speaking rates to features in normal speaking rates. The feature space maximum likelihood linear regression...
Speech emotion recognition is a still challenging problem despite having been investigated over the last couple of decades. Conventional speech emotion recognition performance is low, but this may be improved by considering new features and an annotation method. In this paper, firstly we use glottal features for speech emotion recognition to improve its performance because the emotions are related...
Virtual military training systems have received considerable attention as a possible substitute for conventional real military training. In our previous work, human action recognition system using multiple Kinects (HARS-MK) has been implemented as a prototype of virtual military training simulator. However, the classification accuracy of HARS-MK is not enough to be utilized for virtual military training...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.