Search

Recent advances in Deep Learning (DL) for speaker recognition have improved the performance but are constrained to the need of labels for the background data, which is difficult in prac- tice. In i-vector based speaker recognition, cosine (unsuper- vised) and PLDA (supervised) are the basic scoring techniques, with a big performance gap between the two. In this thesis we tried to fill this gap without using speaker labels in several ways. We applied Restricted Boltzmann Machine (RBM) vectors for the tasks of speaker clustering and tracking in TV broadcast shows. The experiments on AGORA database show that us- ing this approach we gain a relative improvement of 12% and 11% for speaker clustering and tracking tasks, respectively. We also applied DL techniques in order to increase the discrimina- tive power of i-vectors in speaker verification task, for which we have proposed the use of autoencoder in several ways, i.e., (1) as a pre-training for a Deep Neural Network (DNN), (2) as a near- est neighbor autoencoder for i-vectors, (3) as an average pooled nearest neighbor autoencoder. The experiments on VoxCeleb database show that we gain a relative improvement of 21%, 42% and 53%, using the three systems respectively. Finally we also proposed a self-supervised end-to-end speaker verification system. The architecture is based on a Convolutional Neural Network (CNN), trained as a siamese network with multiple branches. From the results we can see that our system shows comparable performance to a supervised baseline This work was supported by the project PID2019-107579RBI00 / AEI / 10.13039/501100011033

Most state-of-the-art Deep Learning systems for text-independent speaker verification are based on speaker embedding extractors. These architectures are commonly composed of a feature extractor front-end together with a pooling layer to encode variable-length utterances into fixed-length speaker vectors. In this paper we present Double Multi-Head Attention (MHA) pooling, which extends our previous approach based on Self MHA. An additional self attention layer is added to the pooling layer that summarizes the context vectors produced by MHA into a unique speaker representation. This method enhances the pooling mechanism by giving weights to the information captured for each head and it results in creating more discriminative speaker embeddings. We have evaluated our approach with the VoxCeleb2 dataset. Our results show 6.09% and 5.23% relative improvement in terms of EER compared to Self Attention pooling and Self MHA, respectively. According to the obtained results, Double MHA has shown to be an excellent approach to efficiently select the most relevant features captured by the CNN-based front-ends from the speech signal. This work was supported by the Spanish project PID2019-107579RBI00 / AEI / 10.13039/501100011033.

Speech enhancement improves recorded voice utterances to eliminate noise that might be impeding their intelligibility or compromising their quality. Typical speech enhancement systems are based on regression approaches that subtract noise or predict clean signals. Most of them do not operate directly on waveforms. In this work, we propose a generative approach to regenerate corrupted signals into a clean version by using generative adversarial networks on the raw signal. We also explore several variations of the proposed system, obtaining insights into proper architectural choices for an adversarially trained, convolutional autoencoder applied to speech. We conduct both objective and subjective evaluations to assess the performance of the proposed method. The former helps us choose among variations and better tune hyperparameters, while the latter is used in a listening experiment with 42 subjects, confirming the effectiveness of the approach in the real world. We also demonstrate the applicability of the approach for more generalized speech enhancement, where we have to regenerate voices from whispered signals.

In the last years, i-vectors followed by cosine or PLDA scoringtechniques were the state-of-the-art approach in speaker veri-fication. PLDA requires labeled background data, and thereexists a significant performance gap between the two scoringtechniques. In this work, we propose to reduce this gap by us-ing an autoencoder to transform i-vector into a new speaker vec-tor representation, which will be referred to as ae-vector. Theautoencoder will be trained to reconstruct neighbor i-vectors in-stead of the same training i-vectors, as usual. These neighbori-vectors will be selected in an unsupervised manner accordingto the highest cosine scores to the training i-vectors. The evalua-tion is performed on the speaker verification trials of VoxCeleb-1 database. The experiments show that our proposed ae-vectorsgain a relative improvement of 42% in terms of EER comparedto the conventional i-vectors using cosine scoring, which fillsthe performance gap between cosine and PLDA scoring tech-niques by 92%, but without using speaker labels

Most state-of-the-art Deep Learning (DL) approaches for speaker recognition work on a short utterance level. Given the speech signal, these algorithms extract a sequence of speaker embeddings from short segments and those are averaged to obtain an utterance level speaker representation. In this work we propose the use of an attention mechanism to obtain a discriminative speaker embedding given non fixed length speech utterances. Our system is based on a Convolutional Neural Network (CNN) that encodes short-term speaker features from the spectrogram and a self multi-head attention model that maps these representations into a long-term speaker embedding. The attention model that we propose produces multiple alignments from different subsegments of the CNN encoded states over the sequence. Hence this mechanism works as a pooling layer which decides the most discriminative features over the sequence to obtain an utterance level representation. We have tested this approach for the verification task for the VoxCeleb1 dataset. The results show that self multi-head attention outperforms both temporal and statistical pooling methods with a 18\% of relative EER. Obtained results show a 58\% relative improvement in EER compared to i-vector+PLDA., 4+1 pages. 4 Figures. Accepted for Interspeech 2009

HighlightsOverview of hybrid machine translation guided by the rule-based paradigm.Overview of hybrid machine translation guided by the corpus-based paradigm.Review of works on machine translation applications with hybrid components. This survey on hybrid machine translation (MT) is motivated by the fact that hybridization techniques have become popular as they attempt to combine the best characteristics of highly advanced pure rule or corpus-based MT approaches. Existing research typically covers either simple or more complex architectures guided by either rule or corpus-based approaches. The goal is to combine the best properties of each type.This survey provides a detailed overview of the modification of the standard rule-based architecture to include statistical knowledge, the introduction of rules in corpus-based approaches, and the hybridization of approaches within this last single category. The principal aim here is to cover the leading research and progress in this field of MT and in several related applications.

Alzheimer’s disease (AD) is the most prevalent form of degenerative dementia; it has a high socio-economic impact in Westerncountries. The purpose of our project is to contribute to earlier diagnosis of AD and allow better estimates of its severity by usingautomatic analysis performed through new biomarkers extracted through non-invasive intelligent methods. The method selectedis based on speech biomarkers derived from the analysis of spontaneous speech (SS). Thus the main goal of the present work isfeature search in SS, aiming at pre-clinical evaluation whose results can be used to select appropriate tests for AD diagnosis. Thefeature set employed in our earlier work offered some hopeful conclusions but failed to capture the nonlinear dynamics of speechthat are present in the speech waveforms. The extra information provided by the nonlinear features could be especially useful whentraining data is limited. In this work, the fractal dimension (FD) of the observed time series is combined with linear parameters inthe feature vector in order to enhance the performance of the original system while controlling the computational cost.© 2014 Elsevier Ltd. All rights reserved.

Conversion from text to speech relies on the accurate mapping from linguistic to acoustic symbol sequences, for which current practice employs recurrent statistical models such as recurrent neural networks. Despite the good performance of such models (in terms of low distortion in the generated speech), their recursive structure with intermediate affine transformations tends to make them slow to train and to sample from. In this work, we explore two different mechanisms that enhance the operational efficiency of recurrent neural networks, and study their performance&ndash, speed trade-off. The first mechanism is based on the quasi-recurrent neural network, where expensive affine transformations are removed from temporal connections and placed only on feed-forward computational directions. The second mechanism includes a module based on the transformer decoder network, designed without recurrent connections but emulating them with attention and positioning codes. Our results show that the proposed decoder networks are competitive in terms of distortion when compared to a recurrent baseline, whilst being significantly faster in terms of CPU and GPU inference time. The best performing model is the one based on the quasi-recurrent mechanism, reaching the same level of naturalness as the recurrent neural network based model with a speedup of 11.2 on CPU and 3.3 on GPU.

Restricted Boltzmann Machines (RBMs) have shown success in both the front-end and backend of speaker verification systems. In this paper, we propose applying RBMs to the front-end for the tasks of speaker clustering and speaker tracking in TV broadcast shows. RBMs are trained to transform utterances into a vector based representation. Because of the lack of data for a test speaker, we propose RBM adaptation to a global model. First, the global model&mdash, which is referred to as universal RBM&mdash, is trained with all the available background data. Then an adapted RBM model is trained with the data of each test speaker. The visible to hidden weight matrices of the adapted models are concatenated along with the bias vectors and are whitened to generate the vector representation of speakers. These vectors, referred to as RBM vectors, were shown to preserve speaker-specific information and are used in the tasks of speaker clustering and speaker tracking. The evaluation was performed on the audio recordings of Catalan TV Broadcast shows. The experimental results show that our proposed speaker clustering system gained up to 12% relative improvement, in terms of Equal Impurity (EI), over the baseline system. On the other hand, in the task of speaker tracking, our system has a relative improvement of 11% and 7% compared to the baseline system using cosine and Probabilistic Linear Discriminant Analysis (PLDA) scoring, respectively.

Literature review on prosody reveals the lack of corpora for prosodic studies in Catalan and Spanish. In this paper, we present a corpus intended to fill this gap. The corpus comprises two distinct data-sets, a news subcorpus and a dialogue subcorpus, the latter containing either conversational or task-oriented speech. More than 25 h were recorded by twenty eight speakers per language. Among these speakers, eight were professional (four radio news broadcasters and four advertising actors). The entire material presented here has been transcribed, aligned with the acoustic signal and prosodically annotated. Two major objectives have guided the design of this project: (i) to offer a wide coverage of representative real-life communicative situations which allow for the characterization of prosody in these two languages; and (ii) to conduct research studies which enable us to contrast the speakers different speaking styles and discursive practices. All material contained in the corpus is provided under a Creative Commons Attribution 3.0 Unported License.

This work presents a study on the suitability of prosodic andacoustic features, with a special focus on i-vectors, in expressivespeech analysis and synthesis. For each utterance of two dif-ferent databases, a laboratory recorded emotional acted speech,and an audiobook, several prosodic and acoustic features are ex-tracted. Among them, i-vectors are built not only on the MFCCbase, but also on F0, power and syllable durations. Then, un-supervised clustering is performed using different feature com-binations. The resulting clusters are evaluated calculating clus-ter entropy for labeled portions of the databases. Additionally,synthetic voices are trained, applying speaker adaptive training,from the clusters built from the audiobook. The voices are eval-uated in a perceptual test where the participants have to edit anaudiobook paragraph using the synthetic voices.The objective results suggest that i-vectors are very use-ful for the audiobook, where different speakers (book charac-ters) are imitated. On the other hand, for the laboratory record-ings, traditional prosodic features outperform i-vectors. Also,a closer analysis of the created clusters suggest that differentspeakers use different prosodic and acoustic means to conveyemotions. The perceptual results suggest that the proposed i-vector based feature combinations can be used for audiobookclustering and voice training.

Nowadays, biometrics is considered as a promising solution in the market of security and personal verification. Applications such as financial transactions, law enforcement or network management security are already benefitting from this technology. Among the different biometric modalities, speaker verification represents an accurate and efficient way of authenticating a person’s identity by analyzing his/her voice. This identification method is especially suitable in real-life scenarios or when a remote recognition over the phone is required. The processing of a signal of voice, in order to extract its unique features, that allows distinguishing an individual to confirm or deny his/ her identity is, usually, a process characterized by a high computational cost. This complexity imposes that many systems, based on microprocessor clocked at hundreds of MHz, are unable to process samples of voice in real-time. This drawback has an important effect, since in general, the response time needed by the biometric system affects its acceptability by users. The design based on FPGA (Field Programmable Gate Arrays) is a suited way to implement systems that require a high computational capability and the resolution of algorithms in real-time. Besides, these devices allow the design of complex digital systems with outstanding performance in terms of execution time. This paper presents the implementation of a MFCC (Mel-Frequency Cepstrum Coefficients)—SVM (Support Vector Machine) speaker verification system based on a low-cost FPGA. Experimental results show that our system is able to verify a person’s identity as fast as a high-performance microprocessor based on a Pentium IV personal computer.

The synthesis of talking heads has been a flourishing research area over the last few years. Since human beings have an uncanny ability to read people's faces, most related applications (e.g., advertising, video-teleconferencing) require absolutely realistic photometric and behavioral synthesis of faces. This paper proposes a person-specific facial synthesis framework that allows high realism and includes a novel way to control visual emphasis (e.g., level of exaggeration of visible articulatory movements of the vocal tract). There are three main contributions: a geodesic interpolation with visual unit selection, a parameterization of visual emphasis, and the design of minimum size corpora. Perceptual tests with human subjects reveal high realism properties, achieving similar perceptual scores as real samples. Furthermore, the visual emphasis level and two communication styles show a statistical interaction relationship.

Deep Learning has been applied successfully to speech processing. In this paper we propose an architecture for speech synthesis using multiple speakers. Some hidden layers are shared by all the speakers, while there is a specific output layer for each speaker. Objective and perceptual experiments prove that this scheme produces much better results in comparison with sin- gle speaker model. Moreover, we also tackle the problem of speaker interpolation by adding a new output layer (a-layer) on top of the multi-output branches. An identifying code is injected into the layer together with acoustic features of many speakers. Experiments show that the a-layer can effectively learn to interpolate the acoustic features between speakers.

This work aims at creating expressive voices from audiobooks using semantic selection. First, for each utterance of the audiobook an acoustic feature vector is extracted, including iVectors built on MFCC and on F0 basis. Then, the transcription is projected into a semantic vector space. A seed utterance is projected to the semantic vector space and the N nearest neighbors are selected. The selection is then filtered by selecting only acoustically similar data. The proposed technique can be used to train emotional voices by using emotional keywords or phrases as seeds, obtaining training data semantically similar to the seed. It can also be used to read larger texts in an expressive manner, creating specific voices for each sentence. That later application is compared to a DNN predictor, which predicts acoustic features from semantic features. The selected data is used to adapt statistical speech synthesis models. The performance of the technique is analyzed objectively and in a perceptive experiment. In the first part of the experiment, subjects clearly show preference for particular expressive voices to synthesize semantically expressive utterances. In the second part, the proposed method is shown to achieve similar or better performance than the DNN based prediction. Copyright © 2016 ISCA.

© 2014 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The frequency diversity exhibited by cyclostationary signals is exploited in this paper. A novel rank-1 frequency-domain representation of a digital waveform is proposed to address the generalized likelihood ratio (GLR) detection of a cyclo-stationary signal with unknown white noise. With the aim of avoiding the well-known sensitivity of cyclostationary-based detectors to frequency-selective fading channels, a parametric channel model based on the coherence bandwidth is adopted and incorporated in the GLR test. The proposed detector outperforms the classical spectral correlation magnitude detectors by exploiting the rank-1 structure of small spectral co-variance matrices

This paper presents the perceptual experiments that were carried out in order to validate the methodology of transforming expressive speech styles using voice quality (VoQ) parameters modelling, along with the well-known prosody (F0, duration, and energy), from a neutral style into a number of expressive ones. The main goal was to validate the usefulness of VoQ in the enhancement of expressive synthetic speech in terms of speech quality and style identification. A harmonic plus noise model (HNM) was used to modify VoQ and prosodic parameters that were extracted from an expressive speech corpus. Perception test results indicated the improvement of obtained expressive speech styles using VoQ modelling along with prosodic characteristics.

Acoustic source localization and sound recognition are common acoustic scene analysis tasks that are usually considered separately. In this paper, a new source localization technique is proposed that works jointly with an acoustic event detection system. Given the identities and the end-points of simultaneous sounds, the proposed technique uses the statistical models of those sounds to compute a likelihood score for each model and for each signal at the output of a set of null-steering beamformers per microphone array. Those scores are subsequently combined to find the MAP-optimal event source positions in the room. Experimental work is reported for a scenario consisting of meeting-room acoustic events, either isolated or overlapped with speech. From the localization results, which are compared with those from the SRP-PHAT technique, it seems that the proposed model-based approach can be an alternative to current techniques for event-based localization. © 2014 IEEE.

The work presented here is part of a larger study to identify novel technologies and biomarkers for early Alzheimer disease (AD) detection and it focuses on evaluating the suitability of a new approach for early AD diagnosis by non-invasive methods. The purpose is to examine in a pilot study the potential of applying intelligent algorithms to speech features obtained from suspected patients in order to contribute to the improvement of diagnosis of AD and its degree of severity. In this sense, Artificial Neural Networks (ANN) have been used for the automatic classification of the two classes (AD and control subjects). Two human issues have been analyzed for feature selection: Spontaneous Speech and Emotional Response. Not only linear features but also non-linear ones, such as Fractal Dimension, have been explored. The approach is non invasive, low cost and without any side effects. Obtained experimental results were very satisfactory and promising for early diagnosis and classification of AD patients. his work has been partially supported by a SAIOTEK from the Basque Government, University of Vic under the research grant R0904, and the Spanish Ministerio de Ciencia e Innovacion TEC2012-38630-C04-03. Iciar Martinez (Research Center for Experimental Marine Biology and Biotechnology-Plentziako Itsas Estazioa (PIE), University of the Basque Country & IKERBASQUE, Basque Foundation for Science

Alzheimer's disease (AD) is the most prevalent form of progressive degenerative dementia; it has a high socio-economic impact in Western countries. The purpose of our project is to contribute to earlier diagnosis of AD and better estimates of its severity by using automatic analysis performed through new biomarkers extracted from non-invasive intelligent methods. The methods selected in this case are speech biomarkers oriented to Spontaneous Speech. Thus the main goal of the present work is feature search in Spontaneous Speech oriented to pre-clinical evaluation for the definition of test for AD diagnosis. Nowadays our feature set offers some hopeful conclusions but fails to capture the nonlinear dynamics of speech that are present in the speech waveforms. The extra information provided by the nonlinear features could be especially useful when training data is scarce. In this work, the Fractal Dimension (FD) of the observed time series is combined with lineal parameters in the feature vector in order to enhance the performance of the original system.

Speech signals in real scenario ambient are usually mixed with some other signals, such as noise. This may interfere with posterior signal processing applied to the signals. In this work, a new technique of data denoising is presented using multivariate Empirical Mode Decomposition. Different SNR ratios are tested in order to study the evolution of the improvement of the recovered data. An improvement of the analyzed data is obtained with all the SNR levels tested.

The automatic analysis of speech to detect affective states may improve the way users interact with electronic devices. However, the analysis only at the acoustic level could be not enough to determine the emotion of a user in a realistic scenario. In this paper we analyzed the spontaneous speech recordings of the FAU Aibo Corpus at the acoustic and linguistic levels to extract two sets of features. The acoustic set was reduced by a greedy procedure selecting the most relevant features to optimize the learning stage. We compared two versions of this greedy selection algorithm by performing the search of the relevant features forwards and backwards. We experimented with three classification approaches: Naïve-Bayes, a support vector machine and a logistic model tree, and two fusion schemes: decision-level fusion, merging the hard-decisions of the acoustic and linguistic classifiers by means of a decision tree; and feature-level fusion, concatenating both sets of features before the learning stage. Despite the low performance achieved by the linguistic data, a dramatic improvement was achieved after its combination with the acoustic information, improving the results achieved by this second modality on its own. The results achieved by the classifiers using the parameters merged at feature level outperformed the classification results of the decision-level fusion scheme, despite the simplicity of the scheme. Moreover, the extremely reduced set of acoustic features obtained by the greedy forward search selection algorithm improved the results provided by the full set.

Wavelet derived codification techniques are widespread used in image codifiers. The wavelet based compression methods are adequate for representing transients. In this paper we explore the use of the discrete wavelet transform analysis of biological signals in order to improve the data compression capability of data coders. The wavelet analysis provides a subband decomposition of any signal, and this enables a lossless or a lossy implementation with the same architecture. The signals could range from speech to sounds or music, but the approach is more orientated to other biosignals like medical signals EEG, ECG or discrete series. Experimental results based on wavelet coefficients quantification, show a lossless compression of 2:1 in all kind of signals, with a fidelity, measured using PSNR, from 79dB to 100dB, and lossy results preserving most of the signal waveform, with a compression ratio from 3:1 to 5:1, with a fidelity from 25dB to 35 dB.

In this paper we show how a nonlinear preprocessing of speech signal -with high noise- based on morphological filters improves the performance of robust algorithms for pitch tracking (RAPT). This result happens for a very simple morphological filter. More sophisticated ones could even improve such results. Mathematical morphology is widely used in image processing in where it has found a great amount of applications. Almost all its formulations derived in the two-dimensional framework are easily reformulated to be adapted to onedimensional context.

This special issue aims to cover some problems related to non-linear and non-conventional speech processing. The origin of this volume is in the ISCA Tutorial and Research Workshop on Non-Linear Speech Processing, NOLISP’09, held at the Universitat de Vic (Catalonia, Spain) on June 25–27, 2009. The series of NOLISP workshops started in 2003 has become a biannual event whose aim is to discuss alternative techniques for speech processing that, in a sense, do not fit into mainstream approaches. A selected choice of papers based on the presentations delivered at NOLISP’09 has given rise to this issue of Cognitive Computation. The papers hereinafter deal with the following topics: What would happen if we were able to overcome the limitations of audiovisual human systems? The answer could be found in the paper by V. Espinosa-Duro, M. Faundez-Zanuy, and J. Mekyska. A growing recurrent self-organizing model for phoneme recognition is presented by Ch. Jlassi, N. Arous, and N. Ellouze. An algorithm for voiced/unvoiced decision and pitch estimation from speech signals based on classification of peaks using the auto-correlation of the speech multi-scale product is presented by M. A. Ben Massoud, A. Bouzid, and N. Elloze. A feature reduction system based on the Discriminative Common Vector is proposed by C.M. Travieso, M. del Pozo, M.A. Ferrer, and J. B. Alonso. A new approach to the diagnosis of Ataxia SCA-2 by application of independent component analysis to a set of data obtained by electro-oculography is presented by R. V. Garcia, F. Rojas, C. G. Puntonet, B. San Roman, L. Velazquez, and R. Rodriguez. A formant tracking technique based on Fourier ridges detection aiming at improving the performance of these algorithms is presented by I. Jemaa, K. Ouni and Y. Laprie A new architecture for vocabulary independent keyword detection for cognitive virtual agents (SEMAINE system) is proposed by M. Wollmer, F. Eyber, A. Graves, B. Schuller, and G. Rigoll. Their word spotting model is composed of a dynamic-Bayesian network and a bidirectional long–short-term memory recurrent neural network. An insight into non-linear transformations for improving the performance of an entropy-based voice activity detector is made in the paper by two guest editors (J. S. and V.Z.). By no means has the range of topics in this Special Issue covered all directions in non-linear speech processing. We would rather consider it a sample presenting several areas in progress. In any case, we hope that they will be of interest for the readers of Cognitive Computation. Our thanks are due to the Editor-in-Chief, Dr. Amir Hussain, for permanent support in preparation of this issue. Of course, this issue would have been impossible without the contributions of all authors. On the other hand, anonymous referees have also done a great job by carefully J. Sole-Casals (&) V. Zaiats Department of Digital Technologies and Information, Escola Politecnica Superior, Universitat de Vic, c/. de la Laura, 13, 08500 Vic (Barcelona), Spain e-mail: jordi.sole@uvic.cat

This paper presents a thorough study of the impact of morphology derivation on N-gram-based Statistical Machine Translation (SMT) models from English into a morphology-rich language such as Spanish. For this purpose, we define a framework under the assumption that a certain degree of morphology-related information is not only being ignored by current statistical translation models, but also has a negative impact on their estimation due to the data sparseness it causes. Moreover, we describe how this information can be decoupled from the standard bilingual N-gram models and introduced separately by means of a well-defined and better informed feature-based classification task. Results are presented for the European Parliament Plenary Sessions (EPPS) English ¿ Spanish task, showing oracle scores based on to what extent SMT models can benefit from simplifying Spanish morphological surface forms for each Part-Of-Speech category. We show that verb form morphological richness greatly weakens the standard statistical models, and we carry out a posterior morphology classification by defining a simple set of features and applying machine learning techniques. In addition to that, we propose a simple technique to deal with Spanish enclitic pronouns. Both techniques are empirically evaluated and final translation results show improvements over the baseline by just dealing with Spanish morphology. In principle, the study is also valid for translation from English into any other Romance language (Portuguese, Catalan, French, Galician, Italian, etc.). The proposed method can be applied to both monotonic and non-monotonic decoding scenarios, thus revealing the interaction between word-order decoding and the proposed morphology simplification techniques. Overall results achieve statistically significant improvement over baseline performance in this demanding task.

This paper describes a multi-domain text-to-speech (MD-TTS)synthesis strategy for generating speech among different domainsand so increasing the flexibility of high quality TTS systems. Tothat effect, the MD-TTS introduces a flexible TTS architecture thatincludes an automatic domain classification module, which allowsMD-TTS systems to be implemented by different synthesis strate-gies and speech corpus typologies. In this work, the performanceof a corpus-based MD-TTS system is subjectively validated bymeans of several perceptual tests

In this work we present a waveform speech coding system including vector quantization. This system can be seen as a vector version of the scalar ADPCM speech coder. In such system the speech samples are grouped in vectors that are coded by a vector quantizer when its prediction subtraction has been made. This prediction is obtained in the coder by a vector predictor. Due to the non-stationarity of the speech signal, the code must be continuously adapted to the local characteristics of the current input signal of the speech. We propose the use of an adaptive vector predictor that follows the speech statistics variations so the prediction error to be coded presents the minimum dynamic range. On the other side it is wellknown that the prediction error is proportional to the signal energy. To compensate this effect a vector quantizer "gain-shape" model has been proposed, so vectors gain and its shape are separately coded. The obtained empirical results are very promising and exhibit good competitivity with other solutions existing in the literature.

We propose an architecture for speaker recognition. This architecture is independent of the text, robust with the presence of noise, and is based on self organizing maps (SOM) (T. Kohonen, 1984). We compare the performance of this architecture for different parametrizations, different signal to noise ratios, with another method for speaker identification based on the arithmetic harmonic spherity measure on covariance matrices (F. Bimbot and L. Mathan, 1993; J. Hernando et al., 1994)

Spanish is a global language, spoken in a big number of different countries with a big dialectal variability‥ This paper deals with the suitability of using a single multi-dialectal acoustic modeling for all the Spanish variants spoken in Europe and Latin America. This paper deals with the suitability of using a single multi-dialectal acoustic modeling for all the Spanish variants spoken in Europe and Latin America. The objective is two fold. First, it allows to use all the available databases to jointly train and improve the same system. Second, it allows to use a single system for all the Spanish speakers. The paper describes the rule- based phonetic transcription used for each dialectal variant, the selection of the shared and the specific phonemes to be modeled in a multi-dialectal recognition system, and the results of a multi-dialectal system dealing with dialects in and out of the training set.

The article presents a robust representation of speech based on AR modeling of the causal part of the autocorrelation sequence. In noisy speech recognition, this new representation achieves better results than several other related techniques.

Speech dynamic features are routinely used in current speech recognition systems in combination with short-term (static) spectral features. Although many existing speech recognition systems do not weight both kinds of features, it seems convenient to use some weighting in order to increase the recognition accuracy of the system. In the cases that this weighting is performed, it is manually tuned or it consists simply in compensating the variances. The aim of this paper is to propose a method to automatically estimate an optimum state-dependent stream weighting in a continuous density hidden Markov model (CDHMM) recognition system by means of a maximum-likelihood based training algorithm. Unlike other works, it is shown that simple constraints on the new weighting parameters permit to apply the maximum-likelihood criterion to this problem. Experimental results in speaker independent digit recognition show an important increase of recognition accuracy.

We study some speech enhancement algorithms based on the iterative Wiener filtering method due to Lim-Oppenheim (1978), where the AR spectral estimation of the speech is carried out using a second-order analysis. But in our algorithms we consider an AR estimation by means of cumulant analysis. This work extends some preceding papers due to the authors, providing a generalization of third- and fourth-order algorithms by means of adding two parameters in the general expression of Wiener filtering. These parameters allow a better control of their performance. Some results are presented considering AWGN but listening tests give similar performance when other noises (diesel engine) are considered.

A procedure for feature selection in isolated word recognition is discussed. The feature selection is performed in two steps. The first step takes into account the temporal correlation among feature vectors in order to obtain a transformation matrix which projects the initial template of N feature vectors to a new space where they are uncorrelated. This step gives a new template of M feature vectors, where M >

Automatic Speech Recognition (ASR) has experienced a dramatic evolution since pioneer development of Bell Lab’s single-digit recognizer more than 50 years ago. Current ASR systems have taken advantage of the tremendous improvements in AI during the past decade by incorporating Deep Neural Networks into the system and pushing their accuracy to levels comparable to that of humans. This article describes and characterizes a representative ASR system with state-of-the-art accuracy and proposes a hardware platform capable of decoding speech in real-time with a power dissipation close to 1 Watt. The software is based on the so-called hybrid approach with a vocabulary of 200K words and RNN-based language model re-scoring, whereas the hardware consists of a commercially available low-power processor along with two accelerators used for the most compute-intensive tasks. The article shows that high performance can be obtained with very low power, enabling the deployment of these systems in extremely power-constrained environments such as mobile and IoT devices. This work has been supported by the CoCoUnit ERC Advanced Grant of the EU’s Horizon 2020 program (grant No. 833057), the Spanish State Research Agency under grant TIN2016-75344-R (AEI/FEDER, EU), the ICREA Academia program, and the Spanish MICINN Ministry under grant BES-2017-080605.

Highlights? We design a steganography model of speech-in-speech signals. ? The model, Efficient Wavelet Masking, is based on the masking property of the HAS. ? The core is the efficient sorting of the wavelet coefficients of the secret message. ? EWM and other methods such as LSB, FM, SS, SSA, have been analyzed. ? Statistical transparency is better in EWM than in LSB, FM, SS and SSA. Steganography is the process of hiding information on a host signal. Transparency is referred to the ability to avoid suspicion about the existence of a secret message. The most popular mechanisms for hiding data in audio signals are the Least Significant Bit (LSB) substitution, Frequency Masking (FM), Spread Spectrum (SS), and Shift Spectrum Algorithm (SSA). In this paper, we adapt the Frequency Masking concept using an efficient sorting of the wavelet coefficients of the secret messages and use an indirect LSB substitution for hiding speech signals into speech signals. The experimental results show that the proposed model, the Efficient Wavelet Masking (EWM) scheme, has a hiding capacity significantly higher than the Spread and Shift Spectrum Algorithms and additionally a statistical transparency higher than all of the above mentioned mechanisms. Moreover, the transparency is not dependent of the host signal chosen because the wavelet sorting guarantees the adaptation of the secret message to the host signal.

The purpose of our project is to contribute to earlier diagnosis of AD and better estimates of its severity by using automatic analysis performed through new biomarkers extracted from non-invasive intelligent methods. The methods selected in this case are speech biomarkers oriented to Sponta-neous Speech and Emotional Response Analysis. Thus the main goal of the present work is feature search in Spontaneous Speech oriented to pre-clinical evaluation for the definition of test for AD diagnosis by One-class classifier. One-class classifi-cation problem differs from multi-class classifier in one essen-tial aspect. In one-class classification it is assumed that only information of one of the classes, the target class, is available. In this work we explore the problem of imbalanced datasets that is particularly crucial in applications where the goal is to maximize recognition of the minority class as in medical diag-nosis. The use of information about outlier and Fractal Dimen-sion features improves the system performance.

In cognitive radio (CR), the problem of limited spectral resources is solved by enabling unlicensed systems to opportunistically utilize the unused licensed bands. Compressive Sensing (CS) has been successfully applied to alleviate the sampling bottleneck in wideband spectrum sensing leveraging the sparseness of the signal spectrum in open-access networks. This has inspired the design of a number of techniques that identify spectrum holes from sub-Nyquist samples. However, the existence of interference emanating from low-regulated transmissions, which cannot be taken into account in the CS model because of their non-regulated nature, greatly degrades the identification of licensed activity. Capitalizing on the sparsity described by licensed users, this paper introduces a feature-based technique for primary user's spectrum identification with interference immunity which works with a reduced amount of data. The proposed method detects which channels are occupied by primary users' and also identify the primary users transmission powers without ever reconstructing the signals involved. Simulation results show the effectiveness of the proposed technique for interference suppression and primary user detection.

A continuous speech recognition system (called RAMSES) has been built based on the demisyllable as phonetic unit and tools from connected speech recognition. Speech is parameterized by band-pass lifted LPC-cepstra and demisyllables are represented by hidden Markov models (HMM). In this paper, the application of this system to recognize integer numbers from zero to one thousand is described. The paper contains a general overview of the system, an outline of the grammar inference, a description of the HMM training procedure and an assessment on the recognition performance in a speaker independent experiment.

In this article, we present the evaluation results for the task of speaker diarization of broadcast news, which was part of the Albayzin 2010 evaluation campaign of language and speech technologies. The evaluation data consists of a subset of the Catalan broadcast news database recorded from the 3/24 TV channel. The description of five submitted systems from five different research labs is given, marking the common as well as the distinctive system features. The diarization performance is analyzed in the context of the diarization error rate, the number of detected speakers and also the acoustic background conditions. An effort is also made to put the achieved results in relation to the particular system design features.

Several business-to-business and business-to-consumer services are provided as a human-to-human conversation in which the provider representative guides the conversation towards its resolution based on her experience, following internal guidelines. Several attempts to automatize these services are becoming popular, but they are currently limited to procedures and objectives set during design step. Process discovery techniques could provide the necessary mechanisms to monitor event logs derived from textual conversations and expand the capabilities of conversational bots. Still, variability of textual messages hinders the utility of process discovery techniques by producing non-understandable unstructured process models. In this paper, we propose the usage of word embedding for combining events that have a semantically similar name.