Artificial Neural Network based Automatic Speech Recognition Engine for Voice Controlled Micro Air Vehicles

The Relative Distance Vector Neural Network (RDVNN) Model: A Hybrid Approach to Speech Recognition

In this paper we investigated Artificial Neural Networks (ANN) based Automatic Speech Recognition (ASR) by using limited Arabic vocabulary corpora. These limited Arabic vocabulary subsets are digits and vowels carried by specific carrier words. In addition to this, Hidden Markov Model (HMM) based ASR systems are designed and compared to two ANN based systems, namely Multilayer Perceptron (MLP) and recurrent architectures, by using the same corpora. All systems are isolated word speech recognizers. The ANN based recognition system achieved 99.5% correct digit recognition. On the other hand, the HMM based recognition system achieved 98.1% correct digit recognition. With vowels carrier words, the MLP and recurrent ANN based recognition systems achieved 92.13% and 98.06, respectively, correct vowel recognition; but the HMM based recognition system achieved 91.6% correct vowel recognition.

Arabic language can be used by native and non- native speakers; due to Arabic is the language of the holy book of Muslims. In this paper, Arabic phoneme recognition system is proposed based on Malay speakers. This system consists of three main stages. The first stage is noise reduction and it aims to enhance the phoneme signals by excluding the unvoiced signals and keep only the voiced signal. Wiener filter is adapted to accomplish this task. The second stage is based on Mel-Frequency Cepstral Coefficients method to extract a vector of features to represent each phoneme signal. Eventually, pattern recognition neural network is designed as recognizer. The proposed system produces sufficient outcomes with 20 hidden neurons. Keywords—Arabic; Malay; pattern recognition; Wiener; Mel- Frequency Cepstral Coefficients I. INTRODUCTION It is well known the wide range of Artificial Neural Networks (ANN) application in speech recognition, financial, telecommunications, electronics and medical. As for speech recognition system for Arabic language, (1) & (2) have addressed the implementation of recognition system based on Arabic native speakers. In general, neural networks (NNs) for phonemes recognition are divided into several categories for instance Very Large Scale Integration (VLSI) NN that can be divided into digital and analogue with digital NNs being more compatible with feed-forward neural networks while analogue NNs are found to be more successful with recurrent NNs (3). From Automatic Speech System (ASR) point of view, NNs for recognition purpose can be divided into two main category namely conventional neural networks such as Multilayer Perceptron (MLP) and Radial Basis Function (RBF) and secondly Recurrent Neural Networks (RNN). The first category of NNs was implemented as pattern classifiers and proven capable for recognition purpose but not at par as compared to Hidden Markov Model (HMM) performance (2). At present, HMM has been proven to be the most successful approach in ASR research area until recently the possibility to hybrid both HMM and ANN (4) & (5). In general, speech recognition systems consist of three stages specifically noise reduction, feature extraction and classification. The aim of noise reduction is to reduce the level of noise in the speech signal and make it more compatible for the ASR. As for feature extraction, this process attempted to extract the most relevant features related to the phoneme signal. The extracted features acted as input features of the NN during training and testing. Finally, the purpose of classification process is to arrange the most similar and related features in categories based on the training phase. Basically, networks training database are being observed and updated during training session of the networks, involving the weights and biases arguments. Networks with lower errors can be considered as better networks (6). Hence, this paper deems to explore further the ability of NN for Arabic phonemes recognition spoken by Malay individuals. This paper is arranged as following; related work is discussed in section II followed by section III that discussed method to be implemented based on Wiener filter, Mel-Frequency Cepstral Coefficients and pattern recognition neural network. Section IV elaborated experimental analysis and results and followed by Conclusion section.

The movements of talkers' face, nose, mouth and throat are known to convey visual cues and represent several different kinds of informationl, and that can improve speech recognition rate, especially for persons with speech-impairments. We proposed a new speech recognition method using these visual features and hidden Markov model (HMM). Based on global optimisation, a new genetic algorithm (GA) for training HMM was proposed. Six chinese vowels were taken as the experimental data, ten handicapped speakers were taken as the testee. Recognition experiments show that the method is effective and high speed and accuracy for speech recognition. At present, the average recognition rate is 91.47% using improved HMM and 88.96% using the classic training HMM algorithm, So the features has very good robustness and the improved HMM is very good.

In this paper we investigated Artificial Neural Networks (ANN) based Automatic Speech Recognition (ASR) by using limited Arabic vocabulary corpora. These limited Arabic vocabulary subsets are digits and vowels carried by specific carrier words. In addition to this, Hidden Markov Model (HMM) based ASR systems are designed and compared to ANN based systems by using the same corpora. All systems are isolated word speech recognizers. The ANN based recognition system achieved 99.5% correct digit recognition. On the other hand, the HMM based recognition system achieved 98.1% correct digit recognition. With vowels carrier words, the ANN based recognition system achieved 92.13% correct vowel recognition; but the HMM based recognition system achieved 91.6% correct vowel recognition.

Among speech researchers, it is widely believed that Hidden Markov Models (HMMs) are the most successful modelling approaches for acoustic events in speech recognition. However, common assumptions limit the classification abilities of HMMs and these can been relaxed by introducing neural networks in the HMM framework. With today's advances in VLSI technology, artificial neural networks (ANNs) can be integrated into a single chip offering adequate circuit complexity required to attain both a high recognition accuracy and an improved learning time. Analogue implementations are considered due to the high processing speeds. The relative performance of different speech coding parameters for use with two different ANN architectures that lend themselves to analogue hardware implementations are investigated. In this case, the dynamic ranges of the different coefficients need to be taken into consideration since they will affect the performance of the analogue chip due to the scaling of the coefficients to voltage signals. The hardware requirements for implementing the two architectures are then discussed.

The performance of automatic speech recognition (ASR) may be degraded when accented speech is recognized because the speech has some linguistic differences from standard speech. Conventional accented speech recognition studies have utilized the accent embedding method, in which the accent embedding features are directly fed into the ASR network. Although the method improves the performance of accented speech recognition, it has some restrictions, such as increasing the computational costs. This study proposes an efficient method of training the ASR model for accented speech in a domain adversarial way based on the Domain Adversarial Neural Network (DANN). The DANN plays a role as a domain adaptation in which the training data and test data have different distributions. Thus, our approach is expected to construct a reliable ASR model for accented speech by reducing the distribution differences between accented speech and standard speech. DANN has three sub-networks: the feature extractor, the domain classifier, and the label predictor. To adjust the DANN for accented speech recognition, we constructed these three sub-networks independently, considering the characteristics of accented speech. In particular, we used an end-to-end framework based on Connectionist Temporal Classification (CTC) to develop the label predictor, a very important module that directly affects ASR results. To verify the efficiency of the proposed approach, we conducted several experiments of accented speech recognition for four English accents including Australian, Canadian, British (England), and Indian accents. The experimental results showed that the proposed DANN-based model outperformed the baseline model for all accents, indicating that the end-to-end domain adversarial training effectively reduced the distribution differences between accented speech and standard speech.

MAVs (micro air vehicles) with a maximal dimension of 15 cm and nominal flight speeds of around 10 m s−1, operate in a Reynolds number regime of 105 or lower, in which most natural flyers including insects, bats and birds fly. Furthermore, due to their light weight and low flight speed, the MAVs' flight characteristics are substantially affected by environmental factors such as wind gust. Like natural flyers, the wing structures of MAVs are often flexible and tend to deform during flight. Consequently, the aero/fluid and structural dynamics of these flyers are closely linked to each other, making the entire flight vehicle difficult to analyze. We have recently developed a hummingbird-inspired, flapping flexible wing MAV with a weight of 2.4–3.0 g and a wingspan of 10–12 cm. In this study, we carry out an integrated study of the flexible wing aerodynamics of this flapping MAV by combining an in-house computational fluid dynamic (CFD) method and wind tunnel experiments. A CFD model that has a realistic wing planform and can mimic realistic flexible wing kinematics is established, which provides a quantitative prediction of unsteady aerodynamics of the four-winged MAV in terms of vortex and wake structures and their relationship with aerodynamic force generation. Wind tunnel experiments further confirm the effectiveness of the clap and fling mechanism employed in this bio-inspired MAV as well as the importance of the wing flexibility in designing small flapping-wing MAVs.

Micro Air Vehicles (MAVs) have become more attractive for various missions including surveillance or reconnaissance in recent years. MAVs should be capable of maintaining their attitudes through either inherent passive stability or active feedback in order to successfully perform their directives. Stability and Controllability Augmentation Systems (SCASs) are usually employed to enhance the flight performance of conventional aircrafts and Unmanned Aerial Vehicles (UAVs). However, it is no simple task to obtain an accurate numerical model for the flight dynamics of a MAV. An alternative approach for SCASs would be to incorporate reinforcement learning in order to address this numerical complexity. Such implementation has already been successful in other vehicles, such as unmanned ground vehicles (UGVs), because of their bettered stability compared to aerial vehicles. However, in order to train MAVs to learn how to fly, they must first be airborne. Similar to teaching infants how to walk, this paper presents a new method to provide an effective environment where a MAV can learn how to fly. A test setup was constructed to enable magnetic levitation of a MAV embedded with a permanent magnet. This apparatus allows for flexible experimentation: the position and the altitude of the MAV, the constraint forces, and the resulting moments are all adjustable and fixable. This ‘Pseudo Flight Environment’ was demonstrated with a fixed wing MAV model. In order for the model to maintain a constant altitude, a height hold control system was devised and implemented.

Missions envisioned for Micro Air Vehicles (MAVs) such as Intelligence, Surveillance and Reconnaissance (ISR) in urban environments require exceptional agility. These missions typically involve ight through or in close proximity to buildings, tunnels and trees, and require the MAVs to maneuver using sharp dives and climbs along with small-radius turns. The small size of MAVs allows such ight operations within con ned space but the control e ectors must provide su cient maneuverability, while maintaining stability, with only limited sensing capability onboard the platform. To meet these challenges, researchers have long been attracted by the amazing attributes of biological systems, such as those exhibited by birds and insects. Birds can y in dense ocks, executing rapid maneuvers with g-loads far in excess of modern ghter aircrafts, and yet never collide with each other, despite the absence of air tra c controllers. This paper introduces a novel framework for the design and control of a Micro Air Vehicle. The vehicle's conceptual design is based on biologically-inspired principles and emulates a dragon y (Odonata-Anisoptera). The author is borrowing ight mechanism features from the dragon y and merging them with indigenous designs in creating a novel version of a Flapping Wing MAV, called the QV. The QV design incorporates the simplicity of mechanical construction and the complexity of a sophisticated multi-layered Hybrid & Linear / Non-Linear controller to achieve extended ight times and improved agility compared to other Rotary and Flapping Wing MAV designs. The paper talks about the design and control features of the proposed QV design and gives an overview on the developmental e orts been made towards the prototyping of the QV yer. The potential applications for such a high endurance vehicle are numerous, including air-deployable mass surveillance in cluster and swarm formations. The disposability of the vehicle would help in battleeld deployment as well, where such a MAV would be made available to soldiers for proximity sensing and threat level assessment. Other applications would include search and rescue operations and civilian law-enforcement.

In this dissertation, a novel approach of integrating template matching with statistical modeling is proposed to improve continuous speech recognition. Hidden Markov Modeling (HMMs) has been the dominant approach in statistical speech recognition since it provides a principled way of jointly modeling speech spectral variations and time dynamics. However, HMMs have the shortcoming of assuming the observations being independent within each state, which makes it ineffective in modeling the details of speech temporal evolutions that are important for characterizing nonstationary speech sounds. Template-based methods make comparisons between a test pattern and the templates derived from training data, and therefore they are able to capture speech dynamics and time correlation of speech frames better than HMM based methods. However, template matching requires large memory space and computational time since feature vectors of training data need to be stored in computer memory for access at the recognition stage, which is difficult in large vocabulary continuous speech recognition (LVCSR). Our proposed approach takes advantages of both statistical modeling and template matching, which overcomes the weakness of conventional template-based method and is feasible for LVCSR. We use multiple Gaussian Mixture Model (GMM) indices to represent each frame of speech templates, and define the template unit to be context-dependent phone segments (triphone context). We also use phonetic decision trees borrowed from those commonly used in HMMs to tie triphone templates and predict triphones unseen in training data. Two local distances, log likelihood ratio (LLR) and Kullback-Leibler (KL) divergence, are proposed for dynamic time warping (DTW) based template matching. In order to reduce computational complexity and storage space, we propose methods of minimum distance template selection (MDTS) and maximum log-likelihood template selection (MLTS), and investigate a template compression method on top of template selection to further improve recognition performance. The template based methods were used to rescore lattices generated by baseline HMMs on the tasks of TIMIT continuous phone recognition and teleheath LVCSR and experimental results demonstrated that the proposed approach of integrating template matching with statistical modeling significantly improved recognition performances over the HMM baselines. The template selection methods also provided significant recognition accuracy improvements over the HMM baseline while largely reducing the computation and storage complexities. When all templates or MDTS were used, using the LLR local distance obtained better recognition performance than the KL divergence local distance. For MLTS and template compression, KL divergence local distance provided better performance than the LLR local distance, and the template compression method made further improvements over KL based MLTS. Since the templates were constructed based on the GMM indices extracted from HMM baselines, we also validated the effectiveness of the proposed template methods based on enhanced HMM baselines. Experimental results showed that LLR based all template method was able to consistently improve TIMIT phone recognition accuracies based on four enhanced HMM baselines. Prosodic features such as duration, energy, and pitch can reflect longer span information of speech than conventional single frame vectors but they have commonly been ignored by HMMs. Template based methods provide possibilities to conveniently integrate prosodic features into speech recognition, which has not been well studied in the past. In this dissertation, we investigate combining template based methods with the speech prosodic features of duration, energy and pitch to further improve speech recognition accuracy. The scores of prosodic information were computed by a GMM based method and a non-parametric method, and the prosodic scores were combined with the acoustic scores in triphone template matching. Experimental results obtained on the telehealth task showed that prosodic information had positive effects on vowel sound recognition.

This paper gives details of the development of a speech recognition system for voice activated Ground Control Station (GCS). The speech recognition is implemented using MATLAB and the results are validated against the Hidden Markov Model Tool Kit (HTK), an open source tool for speech recognition. The menu items of Mission planner, a typical open source GCS used for flying of Micro Air Vehicles (MAV) are used for the experiments.

In this paper, a brief overview derived out of detailed survey of speech recognition works reported in Indian languages is described. Robustness of speech recognition systems toward language variation is the recent trend of research in speech recognition technology. To develop a system which can communicate with human in any language like any other human is the foremost requirement in order to design appropriate speech recognition technology for one to all. India is a country which has vast linguistic variations among its billion plus population. Therefore, it provides a sound area of research toward language-specific speech recognition technology. From the beginning of the commercial availability of the speech recognition system, the technology has been dominated by the hidden Markov model (HMM) methodology due to its capability of modeling temporal structures of speech and encoding them as a sequence of spectral vectors. Most of the work done in Indian languages also uses HMM technology. However, from the last 10–15 years after the acceptance of neurocomputing as an alternative to HMM, artificial neural network (ANN)-based methodologies have started to receive attention for application in speech recognition. This is a trend worldwide as part of which few works have also been reported by a few researchers.

We designed a X-type wing of flapping-wing Micro Air Vehicles (MAV) and doing some aerodynamic force testing by the flapping-wing MAV aerodynamic force testing device which designed by ourselves. In order to study the relationship of aerodynamic force on the flapping-wing models, we do these tests of different flapping-wing models in different angles of attack and flapping frequency. Found that X-wings flapping wing MAV rely mainly on aerodynamic wings folded back when the compressed gas emitted and received; increase the X-wings flapping wing MAV flapping frequency can significantly increase aerodynamic wings; and X-flapping wing MAV Aerodynamic wing root shoot ratio and aspect ratio with the decrease, inversely proportional to the relationship.

Micro Air Vehicles (MAVs) are a new type of aircraft maturing day by day and have reached unprecedented levels of growth recently. Similarly to larger Unmanned Air Vehicles (UAVs), MAVs have enormous potential in applications, both military and civilian, like reconnaissance over battlefields and surveillance of urban areas, data relay, air sampling etc. This article describes the development and selection of a fixed wing MAV with the analysis of simulated results. High wing theory with NACA 4412 aerofoil’s analytical data has been used to practically predict the performance of the MAV. KEY WORDS: Fixed wing; MAV; NACA 4412; Clark Y. DOI: http://dx.doi.org/10.3329/mist.v3i0.8048