Machine learning methods are growing in relevance for biometrics and personal information processing in domains such as forensics, e-health, recruitment, and e-learning. In these domains, white-box (human-readable) explanations of systems built on ma
chine learning methods can become crucial. Inductive Logic Programming (ILP) is a subfield of symbolic AI aimed to automatically learn declarative theories about the process of data. Learning from Interpretation Transition (LFIT) is an ILP technique that can learn a propositional logic theory equivalent to a given black-box system (under certain conditions). The present work takes a first step to a general methodology to incorporate accurate declarative explanations to classic machine learning by checking the viability of LFIT in a specific AI application scenario: fair recruitment based on an automatic tool generated with machine learning methods for ranking Curricula Vitae that incorporates soft biometric information (gender and ethnicity). We show the expressiveness of LFIT for this specific problem and propose a scheme that can be applicable to other domains.
The performance of speaker verification systems degrades when vocal effort conditions between enrollment and test (e.g., shouted vs. normal speech) are different. This is a potential situation in non-cooperative speaker verification tasks. In this pa
per, we present a study on different methods for linear compensation of embeddings making use of Gaussian mixture models to cluster shouted and normal speech domains. These compensation techniques are borrowed from the area of robustness for automatic speech recognition and, in this work, we apply them to compensate the mismatch between shouted and normal conditions in speaker verification. Before compensation, shouted condition is automatically detected by means of logistic regression. The process is computationally light and it is performed in the back-end of an x-vector system. Experimental results show that applying the proposed approach in the presence of vocal effort mismatch yields up to 13.8% equal error rate relative improvement with respect to a system that applies neither shouted speech detection nor compensation.
This paper explores two techniques to improve the performance of text-dependent speaker verification systems based on deep neural networks. Firstly, we propose a general alignment mechanism to keep the temporal structure of each phrase and obtain a s
upervector with the speaker and phrase information, since both are relevant for a text-dependent verification. As we show, it is possible to use different alignment techniques to replace the global average pooling providing significant gains in performance. Moreover, we also present a novel back-end approach to train a neural network for detection tasks by optimizing the Area Under the Curve (AUC) as an alternative to the usual triplet loss function, so the system is end-to-end, with a cost function close to our desired measure of performance. As we can see in the experimental section, this approach improves the system performance, since our triplet neural network based on an approximation of the AUC (aAUC) learns how to discriminate between pairs of examples from the same identity and pairs of different identities. The different alignment techniques to produce supervectors in addition to the new back-end approach were tested on the RSR2015-Part I database for text-dependent speaker verification, providing competitive results compared to similar size networks using the global average pooling to extract supervectors and using a simple back-end or triplet loss training.
In this paper we propose a method to model speaker and session variability and able to generate likelihood ratios using neural networks in an end-to-end phrase dependent speaker verification system. As in Joint Factor Analysis, the model uses tied hi
dden variables to model speaker and session variability and a MAP adaptation of some of the parameters of the model. In the training procedure our method jointly estimates the network parameters and the values of the speaker and channel hidden variables. This is done in a two-step backpropagation algorithm, first the network weights and factor loading matrices are updated and then the hidden variables, whose gradients are calculated by aggregating the corresponding speaker or session frames, since these hidden variables are tied. The last layer of the network is defined as a linear regression probabilistic model whose inputs are the previous layer outputs. This choice has the advantage that it produces likelihoods and additionally it can be adapted during the enrolment using MAP without the need of a gradient optimization. The decisions are made based on the ratio of the output likelihoods of two neural network models, speaker adapted and universal background model. The method was evaluated on the RSR2015 database.
In this paper, we propose a new differentiable neural network alignment mechanism for text-dependent speaker verification which uses alignment models to produce a supervector representation of an utterance. Unlike previous works with similar approach
es, we do not extract the embedding of an utterance from the mean reduction of the temporal dimension. Our system replaces the mean by a phrase alignment model to keep the temporal structure of each phrase which is relevant in this application since the phonetic information is part of the identity in the verification task. Moreover, we can apply a convolutional neural network as front-end, and thanks to the alignment process being differentiable, we can train the whole network to produce a supervector for each utterance which will be discriminative with respect to the speaker and the phrase simultaneously. As we show, this choice has the advantage that the supervector encodes the phrase and speaker information providing good performance in text-dependent speaker verification tasks. In this work, the process of verification is performed using a basic similarity metric, due to simplicity, compared to other more elaborate models that are commonly used. The new model using alignment to produce supervectors was tested on the RSR2015-Part I database for text-dependent speaker verification, providing competitive results compared to similar size networks using the mean to extract embeddings.
In a previous work, the authors proposed a Grammatical Evolution algorithm to automatically generate Lindenmayer Systems which represent fractal curves with a pre-determined fractal dimension. This paper gives strong statistical evidence that the pro
bability distributions of the execution time of that algorithm exhibits a heavy tail with an hyperbolic probability decay for long executions, which explains the erratic performance of different executions of the algorithm. Three different restart strategies have been incorporated in the algorithm to mitigate the problems associated to heavy tail distributions: the first assumes full knowledge of the execution time probability distribution, the second and third assume no knowledge. These strategies exploit the fact that the probability of finding a solution in short executions is non-negligible and yield a severe reduction, both in the expected execution time (up to one order of magnitude) and in its variance, which is reduced from an infinite to a finite value.
This paper has been withdrawn by the authors due to a major rewriting.
