In this paper, we propose a new pooling method called spatial pyramid encoding (SPE) to generate speaker embeddings for text-independent speaker verification. We first partition the output feature maps from a deep residual network (ResNet) into increasingly fine sub-regions and extract speaker embeddings from each sub-region through a learnable dictionary encoding layer. These embeddings are concatenated to obtain the final speaker representation. The SPE layer not only generates a fixed-dimensional speaker embedding for a variable-length speech segment, but also aggregates the information of feature distribution from multi-level temporal bins. Furthermore, we apply deep length normalization by augmenting the loss function with ring loss. By applying ring loss, the network gradually learns to normalize the speaker embeddings using model weights themselves while preserving convexity, leading to more robust speaker embeddings. Experiments on the VoxCeleb1 dataset show that the proposed system using the SPE layer and ring loss-based deep length normalization outperforms both ivector and d-vector baselines. Index Terms: speaker verification, spatial pyramid encoding, learnable dictionary encoding, ring loss, length normalization
d-vector systemsWe can classify d-vector based SV systems according to the loss function used. The first one is based on the softmax loss defined in [23] as the combination of a cross-entropy loss, a softmax function and the last fully connected layer [7,8,24]. In this system, a speaker classifier is trained to classify speakers in the training set. The softmax loss encourages the separability of speaker embeddings. However, the softmax loss is not sufficient to learn the discriminative embedding with a large margin, and more researchers began to explore discriminative loss functions for enhanced generalization ability.
Voice activity detection (VAD) is a challenging task in very low signal-to-noise ratio (SNR) environments. To address this issue, a promising approach is to map noisy speech features to corresponding clean features and to perform VAD using the generated clean features. This can be implemented by concatenating a speech enhancement (SE) and a VAD network, whose parameters are jointly updated. In this paper, we propose denoising variational autoencoder-based (DVAE) speech enhancement in the joint learning framework. Moreover, we feed not only the enhanced feature but also the latent code from the DVAE into the VAD network. We show that the proposed joint learning approach outperforms conventional denoising autoencoder-based joint learning approach.
This paper addresses the problem of recognizing the speech uttered by patients with dysarthria, which is a motor speech disorder impeding the physical production of speech. Patients with dysarthria have articulatory limitation, and therefore, they often have trouble in pronouncing certain sounds, resulting in undesirable phonetic variation. Modern automatic speech recognition systems designed for regular speakers are ineffective for dysarthric sufferers due to the phonetic variation. To capture the phonetic variation, Kullback-Leibler divergence based hidden Markov model (KL-HMM) is adopted, where the emission probability of state is parametrized by a categorical distribution using phoneme posterior probabilities obtained from a deep neural network-based acoustic model. To further reflect speaker-specific phonetic variation patterns, a speaker adaptation method based on a combination of L2 regularization and confusion-reducing regularization which can enhance discriminability between categorical distributions of KL-HMM states while preserving speaker-specific information is proposed. Evaluation of the proposed speaker adaptation method on a database of several hundred words for 30 speakers consisting of 12 mildly dysarthric, 8 moderately dysarthric, and 10 non-dysarthric control speakers showed that the proposed approach significantly outperformed the conventional deep neural network based speaker adapted system on dysarthric as well as non-dysarthric speech.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.