Abstract
In this paper we present a new approach to the problem of isolated vowel recognition in real-time. Language learning and speech therapy are examples of application areas that require real-time biofeedback of acoustic features. As the performance of known approaches usually drops for child speakers, we evaluated different alternatives of feature extraction and pattern recognition techniques, including PCA, LDA, ANN and Bayesian classification. In addition, we studied the explicit inclusion of pitch as a main parameter in both simulation and the real-time feature extraction process. Best results were obtained with our dataset when MFCCs are mapped, using LDA, to a 4-dimensional sub-space that is followed by Bayesian classification. An interactive game was designed that implements the selected real-time vowel recognition technique.

Abstract
This paper presents a Genetic Algorithm approach with Harmonic Structure Evolution for Polyphonic Music Transcription. Automatic Music Transcription is a very complex problem that continues waiting for solutions due to the harmonic complexity of musical sounds. More traditional approaches try to extract the information directly from the audio stream, but by taking into account that a polyphonic audio stream is no more than a combination of several musical notes, music transcription can be addressed as a search space problem where the goal is to find the sequence of notes that best models our audio signal. By taking advantage of the genetic algorithms to explore large search spaces we present a new approach to the music transcription problem. In order to reduce the harmonic overfitting several techniques were used including the encoding of the harmonic structure of the internal synthesizer inside the individual's genotype as a way to evolve towards the instrument played on the original audio signal. The results obtained in polyphonic piano transcriptions show the feasibility of the approach.

Abstract
Audio bandwidth extension has emerged as a key low bit rate coding tool. In continuation with our on going research on audio bandwidth extension, this paper presents new enhancements to Audio Bandwidth Extension Toolkit (ABET). ABET consists of three primary tools Accurate Spectral Replacement (ASR), Fractal Self Similarity Model (FSSM) and Multi-band Temporal Envelope Amplitude Coding (MBTAC) [1],[2],[3]. Additionally we have also introduced a blind bandwidth extension mode into ABET [4]. We discuss several new ideas / improvements to ABET. Specifically enhancements to the blind bandwidth extension architecture which allow it to work with signals with only 3.5-4.0 kHz audio bandwidth are described. We also elaborate on a new tool for efficient coding of time-frequency envelope which cuts the overhead by 0.75-1.0 kbps/channel. We also address a practical issue i.e., the computational complexity and describe a new low decoder complexity mode of ABET.

Abstract
This paper presents the Gene Fragment Competition concept that can be used with Hybrid Genetic Algorithms specially in signal and image processing. Memetic Algorithms have shown great success in real-life problems by adding local search operators to improve the quality of the already achieved "good" solutions during the evolutionary process. Nevertheless these traditional local search operators don't perform well in highly demanding evaluation processes. This stresses the need for a new semi-local non-exhaustive method. Our proposed approach sits as a tradeoff between classical Genetic Algorithms and traditional Memetic Algorithms, performing a quasi-global/quasi-local search by means of gene fragment evaluation and selection. The applicability of this hybrid Genetic Algorithm to the signal processing problem of Polyphonic Music Transcription is shown. The results obtained show the feasibility of the approach.

Abstract
In this paper, an evaluation of several methods allowing the estimation of the Harmonic-to-Noise Ratio (HNR) of sustained vowels was conducted. The HNR estimation methods are mainly based on time, spectral, and cepstral signal representations. An algorithm was implemented for each method and was tested with synthesized voice sounds in order to evaluate their accuracy. Tests were also conducted with real pathological voice sounds in order to evaluate the behaviour of the different methods under real conditions. © 2008 Division of Signal Processin.

Abstract
Vowel recognition is frequently based on Linear Prediction (LP) analysis and formant estimation techniques. However, the performance of these techniques decreases in the case of female or child speech because at high pitch frequencies (F0) the magnitude spectrum is scarcely sampled making formant estimation unreliable. In this paper we describe the implementation of a perceptually motivated concept of vowel recognition that is based on Perceptual Spectral Clusters (PSC) of harmonic partials. PSC based features were evaluated in automatic recognition tests using the Mahalanobis distance and using a data base of five natural Portuguese vowel sounds uttered by 44 speakers, 27 of whom are child speakers. LP based features and Mel-Frequency Cepstral Coefficients (MFCC) were also included in the tests as a reference. Results show that while the recognition performance of PSC features falls between that of LP based features and that of MFCC coefficients, the normalization of PSC features by F0 increases the performance and approaches that of MFCC coefficients. PSC features are not only amenable to a psychophysical interpretation (as LP based features are) but have also the potential to compete with global shape features such as MFCCs.