29 resultados para robust speech recognition
Resumo:
OntoTag - A Linguistic and Ontological Annotation Model Suitable for the Semantic Web
1. INTRODUCTION. LINGUISTIC TOOLS AND ANNOTATIONS: THEIR LIGHTS AND SHADOWS
Computational Linguistics is already a consolidated research area. It builds upon the results of other two major ones, namely Linguistics and Computer Science and Engineering, and it aims at developing computational models of human language (or natural language, as it is termed in this area). Possibly, its most well-known applications are the different tools developed so far for processing human language, such as machine translation systems and speech recognizers or dictation programs.
These tools for processing human language are commonly referred to as linguistic tools. Apart from the examples mentioned above, there are also other types of linguistic tools that perhaps are not so well-known, but on which most of the other applications of Computational Linguistics are built. These other types of linguistic tools comprise POS taggers, natural language parsers and semantic taggers, amongst others. All of them can be termed linguistic annotation tools.
Linguistic annotation tools are important assets. In fact, POS and semantic taggers (and, to a lesser extent, also natural language parsers) have become critical resources for the computer applications that process natural language. Hence, any computer application that has to analyse a text automatically and ‘intelligently’ will include at least a module for POS tagging. The more an application needs to ‘understand’ the meaning of the text it processes, the more linguistic tools and/or modules it will incorporate and integrate.
However, linguistic annotation tools have still some limitations, which can be summarised as follows:
1. Normally, they perform annotations only at a certain linguistic level (that is, Morphology, Syntax, Semantics, etc.).
2. They usually introduce a certain rate of errors and ambiguities when tagging. This error rate ranges from 10 percent up to 50 percent of the units annotated for unrestricted, general texts.
3. Their annotations are most frequently formulated in terms of an annotation schema designed and implemented ad hoc.
A priori, it seems that the interoperation and the integration of several linguistic tools into an appropriate software architecture could most likely solve the limitations stated in (1). Besides, integrating several linguistic annotation tools and making them interoperate could also minimise the limitation stated in (2). Nevertheless, in the latter case, all these tools should produce annotations for a common level, which would have to be combined in order to correct their corresponding errors and inaccuracies. Yet, the limitation stated in (3) prevents both types of integration and interoperation from being easily achieved.
In addition, most high-level annotation tools rely on other lower-level annotation tools and their outputs to generate their own ones. For example, sense-tagging tools (operating at the semantic level) often use POS taggers (operating at a lower level, i.e., the morphosyntactic) to identify the grammatical category of the word or lexical unit they are annotating. Accordingly, if a faulty or inaccurate low-level annotation tool is to be used by other higher-level one in its process, the errors and inaccuracies of the former should be minimised in advance. Otherwise, these errors and inaccuracies would be transferred to (and even magnified in) the annotations of the high-level annotation tool.
Therefore, it would be quite useful to find a way to
(i) correct or, at least, reduce the errors and the inaccuracies of lower-level linguistic tools;
(ii) unify the annotation schemas of different linguistic annotation tools or, more generally speaking, make these tools (as well as their annotations) interoperate.
Clearly, solving (i) and (ii) should ease the automatic annotation of web pages by means of linguistic tools, and their transformation into Semantic Web pages (Berners-Lee, Hendler and Lassila, 2001). Yet, as stated above, (ii) is a type of interoperability problem. There again, ontologies (Gruber, 1993; Borst, 1997) have been successfully applied thus far to solve several interoperability problems. Hence, ontologies should help solve also the problems and limitations of linguistic annotation tools aforementioned.
Thus, to summarise, the main aim of the present work was to combine somehow these separated approaches, mechanisms and tools for annotation from Linguistics and Ontological Engineering (and the Semantic Web) in a sort of hybrid (linguistic and ontological) annotation model, suitable for both areas. This hybrid (semantic) annotation model should (a) benefit from the advances, models, techniques, mechanisms and tools of these two areas; (b) minimise (and even solve, when possible) some of the problems found in each of them; and (c) be suitable for the Semantic Web. The concrete goals that helped attain this aim are presented in the following section.
2. GOALS OF THE PRESENT WORK
As mentioned above, the main goal of this work was to specify a hybrid (that is, linguistically-motivated and ontology-based) model of annotation suitable for the Semantic Web (i.e. it had to produce a semantic annotation of web page contents). This entailed that the tags included in the annotations of the model had to (1) represent linguistic concepts (or linguistic categories, as they are termed in ISO/DCR (2008)), in order for this model to be linguistically-motivated; (2) be ontological terms (i.e., use an ontological vocabulary), in order for the model to be ontology-based; and (3) be structured (linked) as a collection of ontology-based
Resumo:
Detecting user affect automatically during real-time conversation is the main challenge towards our greater aim of infusing social intelligence into a natural-language mixed-initiative High-Fidelity (Hi-Fi) audio control spoken dialog agent. In recent years, studies on affect detection from voice have moved on to using realistic, non-acted data, which is subtler. However, it is more challenging to perceive subtler emotions and this is demonstrated in tasks such as labelling and machine prediction. This paper attempts to address part of this challenge by considering the role of user satisfaction ratings and also conversational/dialog features in discriminating contentment and frustration, two types of emotions that are known to be prevalent within spoken human-computer interaction. However, given the laboratory constraints, users might be positively biased when rating the system, indirectly making the reliability of the satisfaction data questionable. Machine learning experiments were conducted on two datasets, users and annotators, which were then compared in order to assess the reliability of these datasets. Our results indicated that standard classifiers were significantly more successful in discriminating the abovementioned emotions and their intensities (reflected by user satisfaction ratings) from annotator data than from user data. These results corroborated that: first, satisfaction data could be used directly as an alternative target variable to model affect, and that they could be predicted exclusively by dialog features. Second, these were only true when trying to predict the abovementioned emotions using annotator?s data, suggesting that user bias does exist in a laboratory-led evaluation.
Resumo:
This paper describes the language identification (LID) system developed by the Patrol team for the first phase of the DARPA RATS (Robust Automatic Transcription of Speech) program, which seeks to advance state of the art detection capabilities on audio from highly degraded communication channels. We show that techniques originally developed for LID on telephone speech (e.g., for the NIST language recognition evaluations) remain effective on the noisy RATS data, provided that careful consideration is applied when designing the training and development sets. In addition, we show significant improvements from the use of Wiener filtering, neural network based and language dependent i-vector modeling, and fusion.
Resumo:
This paper describes a novel approach to phonotactic LID, where instead of using soft-counts based on phoneme lattices, we use posteriogram to obtain n-gram counts. The high-dimensional vectors of counts are reduced to low-dimensional units for which we adapted the commonly used term i-vectors. The reduction is based on multinomial subspace modeling and is designed to work in the total-variability space. The proposed technique was tested on the NIST 2009 LRE set with better results to a system based on using soft-counts (Cavg on 30s: 3.15% vs 3.43%), and with very good results when fused with an acoustic i-vector LID system (Cavg on 30s acoustic 2.4% vs 1.25%). The proposed technique is also compared with another low dimensional projection system based on PCA. In comparison with the original soft-counts, the proposed technique provides better results, reduces the problems due to sparse counts, and avoids the process of using pruning techniques when creating the lattices.
Resumo:
In order to obtain more human like sounding humanmachine interfaces we must first be able to give them expressive capabilities in the way of emotional and stylistic features so as to closely adequate them to the intended task. If we want to replicate those features it is not enough to merely replicate the prosodic information of fundamental frequency and speaking rhythm. The proposed additional layer is the modification of the glottal model, for which we make use of the GlottHMM parameters. This paper analyzes the viability of such an approach by verifying that the expressive nuances are captured by the aforementioned features, obtaining 95% recognition rates on styled speaking and 82% on emotional speech. Then we evaluate the effect of speaker bias and recording environment on the source modeling in order to quantify possible problems when analyzing multi-speaker databases. Finally we propose a speaking styles separation for Spanish based on prosodic features and check its perceptual significance.
Resumo:
This paper presents a description of our system for the Albayzin 2012 LRE competition. One of the main characteristics of this evaluation was the reduced number of available files for training the system, especially for the empty condition where no training data set was provided but only a development set. In addition, the whole database was created from online videos and around one third of the training data was labeled as noisy files. Our primary system was the fusion of three different i-vector based systems: one acoustic system based on MFCCs, a phonotactic system using trigrams of phone-posteriorgram counts, and another acoustic system based on RPLPs that improved robustness against noise. A contrastive system that included new features based on the glottal source was also presented. Official and postevaluation results for all the conditions using the proposed metrics for the evaluation and the Cavg metric are presented in the paper.
Resumo:
When designing human-machine interfaces it is important to consider not only the bare bones functionality but also the ease of use and accessibility it provides. When talking about voice-based inter- faces, it has been proven that imbuing expressiveness into the synthetic voices increases signi?cantly its perceived naturalness, which in the end is very helpful when building user friendly interfaces. This paper proposes an adaptation based expressiveness transplantation system capable of copying the emotions of a source speaker into any desired target speaker with just a few minutes of read speech and without requiring the record- ing of additional expressive data. This system was evaluated through a perceptual test for 3 speakers showing up to an average of 52% emotion recognition rates relative to the natural voice recognition rates, while at the same time keeping good scores in similarity and naturality.
Resumo:
MFCC coefficients extracted from the power spectral density of speech as a whole, seems to have become the de facto standard in the area of speaker recognition, as demonstrated by its use in almost all systems submitted to the 2013 Speaker Recognition Evaluation (SRE) in Mobile Environment [1], thus relegating to background this component of the recognition systems. However, in this article we will show that selecting the adequate speaker characterization system is as important as the selection of the classifier. To accomplish this we will compare the recognition rates achieved by different recognition systems that relies on the same classifier (GMM-UBM) but connected with different feature extraction systems (based on both classical and biometric parameters). As a result we will show that a gender dependent biometric parameterization with a simple recognition system based on GMM- UBM paradigm provides very competitive or even better recognition rates when compared to more complex classification systems based on classical features
Resumo:
Durante el proceso de producción de voz, los factores anatómicos, fisiológicos o psicosociales del individuo modifican los órganos resonadores, imprimiendo en la voz características particulares. Los sistemas ASR tratan de encontrar los matices característicos de una voz y asociarlos a un individuo o grupo. La edad y sexo de un hablante son factores intrínsecos que están presentes en la voz. Este trabajo intenta diferenciar esas características, aislarlas y usarlas para detectar el género y la edad de un hablante. Para dicho fin, se ha realizado el estudio y análisis de las características basadas en el pulso glótico y el tracto vocal, evitando usar técnicas clásicas (como pitch y sus derivados) debido a las restricciones propias de dichas técnicas. Los resultados finales de nuestro estudio alcanzan casi un 100% en reconocimiento de género mientras en la tarea de reconocimiento de edad el reconocimiento se encuentra alrededor del 80%. Parece ser que la voz queda afectada por el género del hablante y las hormonas, aunque no se aprecie en la audición. ABSTRACT Particular elements of the voice are printed during the speech production process and are related to anatomical and physiological factors of the phonatory system or psychosocial factors acquired by the speaker. ASR systems attempt to find those peculiar nuances of a voice and associate them to an individual or a group. Age and gender are inherent factors to the speaker which may be represented in voice. This work attempts to differentiate those characteristics, isolate them and use them to detect speaker’s gender and age. Features based on glottal pulse and vocal tract are studied and analyzed in order to achieve good results in both tasks. Classical methodologies (such as pitch and derivates) are avoided since the requirements of those techniques may be too restrictive. The final scores achieve almost 100% in gender recognition whereas in age recognition those scores are around 80%. Factors related to the gender and hormones seem to affect the voice although they are not audible.
Resumo:
La cuestión principal abordada en esta tesis doctoral es la mejora de los sistemas biométricos de reconocimiento de personas a partir de la voz, proponiendo el uso de una nueva parametrización, que hemos denominado parametrización biométrica extendida dependiente de género (GDEBP en sus siglas en inglés). No se propone una ruptura completa respecto a los parámetros clásicos sino una nueva forma de utilizarlos y complementarlos. En concreto, proponemos el uso de parámetros diferentes dependiendo del género del locutor, ya que como es bien sabido, la voz masculina y femenina presentan características diferentes que deberán modelarse, por tanto, de diferente manera. Además complementamos los parámetros clásicos utilizados (MFFC extraídos de la señal de voz), con un nuevo conjunto de parámetros extraídos a partir de la deconstrucción de la señal de voz en sus componentes de fuente glótica (más relacionada con el proceso y órganos de fonación y por tanto con características físicas del locutor) y de tracto vocal (más relacionada con la articulación acústica y por tanto con el mensaje emitido). Para verificar la validez de esta propuesta se plantean diversos escenarios, utilizando diferentes bases de datos, para validar que la GDEBP permite generar una descripción más precisa de los locutores que los parámetros MFCC clásicos independientes del género. En concreto se plantean diferentes escenarios de identificación sobre texto restringido y texto independiente utilizando las bases de datos de HESPERIA y ALBAYZIN. El trabajo también se completa con la participación en dos competiciones internacionales de reconocimiento de locutor, NIST SRE (2010 y 2012) y MOBIO 2013. En el primer caso debido a la naturaleza de las bases de datos utilizadas se obtuvieron resultados cercanos al estado del arte, mientras que en el segundo de los casos el sistema presentado obtuvo la mejor tasa de reconocimiento para locutores femeninos. A pesar de que el objetivo principal de esta tesis no es el estudio de sistemas de clasificación, sí ha sido necesario analizar el rendimiento de diferentes sistemas de clasificación, para ver el rendimiento de la parametrización propuesta. En concreto, se ha abordado el uso de sistemas de reconocimiento basados en el paradigma GMM-UBM, supervectores e i-vectors. Los resultados que se presentan confirman que la utilización de características que permitan describir los locutores de manera más precisa es en cierto modo más importante que la elección del sistema de clasificación utilizado por el sistema. En este sentido la parametrización propuesta supone un paso adelante en la mejora de los sistemas de reconocimiento biométrico de personas por la voz, ya que incluso con sistemas de clasificación relativamente simples se consiguen tasas de reconocimiento realmente competitivas. ABSTRACT The main question addressed in this thesis is the improvement of automatic speaker recognition systems, by the introduction of a new front-end module that we have called Gender Dependent Extended Biometric Parameterisation (GDEBP). This front-end do not constitute a complete break with respect to classical parameterisation techniques used in speaker recognition but a new way to obtain these parameters while introducing some complementary ones. Specifically, we propose a gender-dependent parameterisation, since as it is well known male and female voices have different characteristic, and therefore the use of different parameters to model these distinguishing characteristics should provide a better characterisation of speakers. Additionally, we propose the introduction of a new set of biometric parameters extracted from the components which result from the deconstruction of the voice into its glottal source estimate (close related to the phonation process and the involved organs, and therefore the physical characteristics of the speaker) and vocal tract estimate (close related to acoustic articulation and therefore to the spoken message). These biometric parameters constitute a complement to the classical MFCC extracted from the power spectral density of speech as a whole. In order to check the validity of this proposal we establish different practical scenarios, using different databases, so we can conclude that a GDEBP generates a more accurate description of speakers than classical approaches based on gender-independent MFCC. Specifically, we propose scenarios based on text-constrain and text-independent test using HESPERIA and ALBAYZIN databases. This work is also completed with the participation in two international speaker recognition evaluations: NIST SRE (2010 and 2012) and MOBIO 2013, with diverse results. In the first case, due to the nature of the NIST databases, we obtain results closed to state-of-the-art although confirming our hypothesis, whereas in the MOBIO SRE we obtain the best simple system performance for female speakers. Although the study of classification systems is beyond the scope of this thesis, we found it necessary to analise the performance of different classification systems, in order to verify the effect of them on the propose parameterisation. In particular, we have addressed the use of speaker recognition systems based on the GMM-UBM paradigm, supervectors and i-vectors. The presented results confirm that the selection of a set of parameters that allows for a more accurate description of the speakers is as important as the selection of the classification method used by the biometric system. In this sense, the proposed parameterisation constitutes a step forward in improving speaker recognition systems, since even when using relatively simple classification systems, really competitive recognition rates are achieved.
Resumo:
This paper presents new techniques with relevant improvements added to the primary system presented by our group to the Albayzin 2012 LRE competition, where the use of any additional corpora for training or optimizing the models was forbidden. In this work, we present the incorporation of an additional phonotactic subsystem based on the use of phone log-likelihood ratio features (PLLR) extracted from different phonotactic recognizers that contributes to improve the accuracy of the system in a 21.4% in terms of Cavg (we also present results for the official metric during the evaluation, Fact). We will present how using these features at the phone state level provides significant improvements, when used together with dimensionality reduction techniques, especially PCA. We have also experimented with applying alternative SDC-like configurations on these PLLR features with additional improvements. Also, we will describe some modifications to the MFCC-based acoustic i-vector system which have also contributed to additional improvements. The final fused system outperformed the baseline in 27.4% in Cavg.
Resumo:
A new language recognition technique based on the application of the philosophy of the Shifted Delta Coefficients (SDC) to phone log-likelihood ratio features (PLLR) is described. The new methodology allows the incorporation of long-span phonetic information at a frame-by-frame level while dealing with the temporal length of each phone unit. The proposed features are used to train an i-vector based system and tested on the Albayzin LRE 2012 dataset. The results show a relative improvement of 33.3% in Cavg in comparison with different state-of-the-art acoustic i-vector based systems. On the other hand, the integration of parallel phone ASR systems where each one is used to generate multiple PLLR coefficients which are stacked together and then projected into a reduced dimension are also presented. Finally, the paper shows how the incorporation of state information from the phone ASR contributes to provide additional improvements and how the fusion with the other acoustic and phonotactic systems provides an important improvement of 25.8% over the system presented during the competition.
Resumo:
A more natural, intuitive, user-friendly, and less intrusive Human–Computer interface for controlling an application by executing hand gestures is presented. For this purpose, a robust vision-based hand-gesture recognition system has been developed, and a new database has been created to test it. The system is divided into three stages: detection, tracking, and recognition. The detection stage searches in every frame of a video sequence potential hand poses using a binary Support Vector Machine classifier and Local Binary Patterns as feature vectors. These detections are employed as input of a tracker to generate a spatio-temporal trajectory of hand poses. Finally, the recognition stage segments a spatio-temporal volume of data using the obtained trajectories, and compute a video descriptor called Volumetric Spatiograms of Local Binary Patterns (VS-LBP), which is delivered to a bank of SVM classifiers to perform the gesture recognition. The VS-LBP is a novel video descriptor that constitutes one of the most important contributions of the paper, which is able to provide much richer spatio-temporal information than other existing approaches in the state of the art with a manageable computational cost. Excellent results have been obtained outperforming other approaches of the state of the art.
Resumo:
The aim of this Master Thesis is the analysis, design and development of a robust and reliable Human-Computer Interaction interface, based on visual hand-gesture recognition. The implementation of the required functions is oriented to the simulation of a classical hardware interaction device: the mouse, by recognizing a specific hand-gesture vocabulary in color video sequences. For this purpose, a prototype of a hand-gesture recognition system has been designed and implemented, which is composed of three stages: detection, tracking and recognition. This system is based on machine learning methods and pattern recognition techniques, which have been integrated together with other image processing approaches to get a high recognition accuracy and a low computational cost. Regarding pattern recongition techniques, several algorithms and strategies have been designed and implemented, which are applicable to color images and video sequences. The design of these algorithms has the purpose of extracting spatial and spatio-temporal features from static and dynamic hand gestures, in order to identify them in a robust and reliable way. Finally, a visual database containing the necessary vocabulary of gestures for interacting with the computer has been created.
