Bruno Gas

Investigation on LP-residual representations for speaker identification

Feature extraction is an essential and important step for speaker recognition systems. In this paper, we propose to improve these systems by exploiting both conventional features such as mel frequency cepstral coding (MFCC), linear predictive cepstral coding (LPCC) and non-conventional ones. The method exploits information present in the linear predictive (LP) residual signal. The features extracted from the LP-residue are then combined to the MFCC or the LPCC. We investigate two approaches termed as temporal and frequential representations. The first one consists of an auto-regressive (AR) modelling of the signal followed by a cepstral transformation in a similar way to the LPC-LPCC transformation. In order to take into account the non-linear nature of the speech signals we used two estimation methods based on second and third-order statistics. They are, respectively, termed as R-SOS-LPCC (residual plus second-order statistic based estimation of the AR model plus cepstral transformation) and R-HOS-LPCC (higher order). Concerning the frequential approach, we exploit a filter bank method called the power difference of spectra in sub-band (PDSS) which measures the spectral flatness over the sub-bands. The resulting features are named R-PDSS. The analysis of these proposed schemes are done over a speaker identification problem with two different databases. The first one is the Gaudi database and contains 49 speakers. The main interest lies in the controlled acquisition conditions: mismatch between the microphones and the interval sessions. The second database is the well-known NTIMIT corpus with 630 speakers. The performances of the features are confirmed over this larger corpus. In addition, we propose to compare traditional features and residual ones by the fusion of recognizers (feature extractor + classifier). The results show that residual features carry speaker-dependent features and the combination with the LPCC or the MFCC shows global improvements in terms of robustness under different mismatches. A comparison between the residual features under the opinion fusion framework gives us useful information about the potential of both temporal and frequential representations.

Discriminant neural predictive coding applied to phoneme recognition

Abstract In this article, we propose to study a speech coding method applied to the recognition of phonemes. The model proposed (the neural predictive coding (NPC) and its three declinations NPC-1, NPC-2 and DFE–NPC) is a connectionist model (multilayer perceptron) based on the nonlinear prediction of the speech signal. We show that it is possible to improve the discriminant capacities of such an encoder with the introduction of signal membership class information as from the coding stage. As such, it fits in with the category of discriminant features extraction (DFE) encoders already proposed in literature. In this study we present a theoretical validation of the model in the hypothesis of unnoised signals and Gaussian noised signals. We also define a new distance, the NPC distance, that will allow experimental validation of the model. NPC performances are compared to that obtained with traditional methods used to process speech on the Darpa Timit phoneme base. Simulations presented here show that the classification rates have clearly improved compared to usual methods, in particular regarding phonemes considered difficult to process (/b/,/d/,/g/ and /p/,/t/,/k/ phonemes).

Optimizing feature complementarity by evolution strategy: Application to automatic speaker verification

Conventional automatic speaker verification systems are based on cepstral features like Mel-scale frequency cepstrum coefficient (MFCC), or linear predictive cepstrum coefficient (LPCC). Recent published works showed that the use of complementary features can significantly improve the system performances. In this paper, we propose to use an evolution strategy to optimize the complementarity of two filter bank based feature extractors. Experiments we made with a state of the art speaker verification system show that significant improvement can be obtained. Compared to the standard MFCC, an equal error rate (EER) improvement of 11.48% and 21.56% was obtained on the 2005 Nist SRE and Ntimit databases, respectively. Furthermore, the obtained filter banks picture out the importance of some specific spectral information for automatic speaker verification.

Binaural Systems in Robotics

Audition is often described by physiologists as the most important sense in humans, due to its essential role in communication and socialization. But quite surprisingly, the interest of this modality for robotics arose only in the 2000s, brought to evidence by cognitive robotics and Human–robot interaction. Since then, numerous contributions have been proposed to the field of robot audition, ranging from sound localization to scene analysis. Binaural approaches were investigated first, then became forsaken due to mixed results. Nevertheless, the last years have witnessed a renewal of interest in binaural active audition, that is, in the opportunities and challenges opened by the coupling of binaural sensing and robot motion. This chapter proposes a comprehensive state of the art of binaural approaches to robot audition. Though the literature on binaural audition and, more generally, on acoustics and signal processing, is a fundamental source of knowledge, the tasks, constraints, and environments of robotics raise original issues. These are reviewed, prior to the most prominent contributions, platforms and projects. Two lines of research in binaural active audition, conducted by the current authors, are then outlined, one of which is tightly connected to psychology of perception.

Self-Organizing MultiLayer Perceptron

In this paper, we propose an extension of a self-organizing map called self-organizing multilayer perceptron (SOMLP) whose purpose is to achieve quantization of spaces of functions. Based on the use of multilayer perceptron networks, SOMLP comprises the unsupervised as well as supervised learning algorithms. We demonstrate that it is possible to use the commonly used vector quantization algorithms (LVQ algorithms) to build new algorithms called functional quantization algorithms (LFQ algorithms). The SOMLP can be used to model nonlinear and/or nonstationary complex dynamic processes, such as speech signals. While most of the functional data analysis (FDA) research is based on B-spline or similar univariate functions, the SOMLP algorithm allows quantization of function with high dimensional input space. As a consequence, classical FDA methods can be outperformed by increasing the dimensionality of the input space of the functions under analysis. Experiments on artificial and real world examples are presented which illustrate the potential of this approach.

Reactive path planning for autonomous sailboat using an omni-directional camera for obstacle detection

Unmanned autonomous vehicles have the potential to operate for extended periods of time in different environments like in the air or the sea. These environments are often complex and arise many challenges for the unmanned vehicles research. The ASAROME project1 (Autonomous Sailing Robot for Oceanographic Measurements) is focused on an autonomous sailboat to make measurements and observations in the marine environment for long term. This paper describes a routing strategy for obstacle avoidance using an omni-directional camera based obstacle detection. Experiments with a panoramic vision system and sailboat simulation results have shown the expected performances for obstacle detection and avoidance.

Filter Bank Design for Speaker Diarization Based on Genetic Algorithms

Speech recognition systems usually need a feature extraction stage aiming at obtaining the best signal representation. In this article we propose to use genetic algorithms to design a feature extraction method adapted to the speaker diarization task. We present an adaptation of the common MFCC feature extractor which consists in designing a filter bank, with optimized bandwidths. Experiments are carried out using a state-of-the-art speaker diarization system. The proposed method outperforms the original filter bank based on the Mel scale one. Furthermore, the obtained filter bank reveals the importance of some specific spectral information for speaker recognition

Non-linear speech feature extraction for phoneme classification and speaker recognition

In this paper we propose a new feature extraction algorithm based on non-linear prediction: the Neural Predictive Coding (NPC) model which is an extension of the classical LPC one. We apply this model to two significant tasks: phoneme classification and speaker identification. For the first one, the NPC model is trained with a Minimum Classification Error (MCE) criterion. The experiments carried out with the NTIMIT database show an improvement of the classification rates. For speaker identification, we propose a new feature extraction principle based on the NPC model. We also investigate different initialization methods. The new method gives better performances than the traditional ones (LPC, MFCC and PLP).

Learning an internal representation of the end-effector configuration space

Current machine learning techniques proposed to automatically discover a robots kinematics usually rely on a priori information about the robots structure, sensor properties or end-effector position. This paper proposes a method to estimate a certain aspect of the forward kinematics model with no such information. An internal representation of the end-effector configuration is generated from unstructured proprioceptive and exteroceptive data flow under very limited assumptions. A mapping from the proprioceptive space to this representational space can then be used to control the robot.

Audio/video fusion for objects recognition

In mobile robotics applications, pattern and object recognition are mainly achieved relying only on vision. Several other perceptual modalities are also available such as, touch, hearing or vestibular proprioception. They are rarely used and can provide valuable additional information within the recognition tasks. This article presents an analysis of several methods of fusion of perceptual and auditory modalitites. It relies on the use of a perspective camera and a microphone on a moving object recognition problem. Experimental data are also provided on a database of audio/visual objects including cases of visual occlusions and audio corruptions.