Auditory-based robust speech recognition system for ambient assisted living in smart home

A statistically designed Automatic Speech Recognition (ASR) system extracts features from speech signals using feature extraction methods, links the extracted features with the expected phonetics of the hypothesis using acoustic models, uses language model to add prior information about the structure of the target language. For many years, Mel-frequency Cepstral coefficients (MFCC), n-gram, and Hidden Markov Model (HMM) approaches have been used predominantly for feature extraction, language modeling and acoustic modeling, respectively. However, performance degradation of MFCC in noisy conditions and inaccuracy of HMMs while handling large vocabularies have made researchers to propose more efficient methods. The proposed work uses noise robust method Gammatone Frequency Cepstral Coefficients(GFCC) for feature extraction, trigram language modeling, and HMM-Gaussian mixture model (GMM) based acoustic modeling to implement a continuous Hindi language ASR system. Also, it applies Differential Evolution (DE) technique to refine the GFCC features and discriminative techniques to enhance performance of the acoustic model. The performance of the implemented system has been evaluated by using different feature extraction methods, variants of n-gram language modeling techniques and different discriminative techniques in clean as well as noisy conditions. Initially, the results reveal that DE optimized GFCC with HMM-Gaussian Mixture Model (GMM) acoustic modeling performs better than MFCC, PLP and MF-PLP feature extraction methods. Secondly, the experimental results show that the Minimum Phone Error (MPE) outperforms Maximum Mutual Information (MMI) and Maximum Likelihood Estimation (MLE) and trigram based language modeling gives more accurate results than unigram and bigram language modeling. Finally, it has been concluded that the continuous Hindi language ASR system implemented using DE refined GFFC feature extraction method with MPE discriminative training technique and trigram based language modeling gives better accuracy in clean as-well-as noisy environments.

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.

In this work, we propose a novel feature extraction algorithm that improves the robustness of automatic speech recognition (ASR) systems in the presence of various types of noise. The proposed algorithm uses a new cepstral technique based on the differential power spectrum (DPS) instead of the power spectrum (PS), the algorithm replaces the logarithmic non linearity by the power function. In order to reduce cepstral coefficients mismatches between training and testing conditions, we used the mean and variance normalization, then we apply auto-regression movingaverage filtering (MVA) in the cepstral domain. The ASR experiments were conducted using two databases, the first is LASA digit database designed for recognition the isolated Arabic digits in the presence of different types of noise. The second is Aurora 2 noisy speech database designed to recognize connected English digits in various operating environments. The experimental results show a substantial improvement from the proposed algorithm over the baseline Mel Frequency Cepstral Coefficients (MFCC), the relative improvement is the 28.92% for LASA database and is the 44.43% for aurora 2 database. The performance of our proposed algorithm was tested and verified by extensive comparisons with the state-of-the-art noise-robust features in aurora 2.

Audio forensics is the application of science and scientific methods in handling digital evidence in the form of audio. In this regard, the audio supports the disclosure of various criminal cases and reveals the necessary information needed in the trial process. So far, research related to audio forensics is more on human voices that are recorded directly, either by using a voice recorder or voice recordings on smartphones, which are available on Google Play services or iOS Store. This study compares the analysis of live voices (human voices) with artificial voices on Google Voice and other artificial voices. This study implements the audio forensic analysis, which involves pitch, formant, and spectrogram as the parameters. Besides, it also analyses the data by using feature extraction using the Mel Frequency Cepstral Coefficient (MFCC) method, the Dynamic Time Warping (DTW) method, and applying the K-Nearest Neighbor (KNN) algorithm. The previously made live voice recording and artificial voice are then cut into words. Then, it tests the chunk from the voice recording. The testing of audio forensic techniques with the Praat application obtained similar words between live and artificial voices and provided 40,74% accuracy of information. While the testing by using the MFCC, DTW, KNN methods with the built systems by using Matlab, obtained similar word information between live voice and artificial voice with an accuracy of 33.33%.

An automatic speech recognition (ASR) is commonly used in these days. Current ASR systems perform well in ideal environment; however, it does not perform well in realistic noisy environment. As a robust ASR, ETSI has standardized Advanced Front-End (AFE) that adopts two-stage of iterative Wiener filter (IWF) to realize a speech enhancement as the front-end of ASR. In the ETSI AFE, 16 kHz speech is divided uniformly by QMF (Quadrature Mirror Filter) into lower-band and higher-band signal and only lower-band signal is emphasized by the IWF and MFCC (Mel Frequency Cepstral Coefficient) is extracted from both the second-stage of emphasized lower-band signal and non-emphasized higher-band signal. FFT is used to estimate speech spectrum that designs the Wiener filter. On the other hand, we have already proposed robust complex speech analysis for an analytic signal. It can estimate more robust and more accurate speech spectrum due to the introduced robust criterion and nature of analytic signal. This paper proposes an improved AFE using wide-band robust ELS (Extended Least Square) complex analysis and real-valued analysis instead of FFT. Moreover, the QMF synthesis is also introduced for second-stage of wide-band analysis. The experimental results using the CENSREC-2 speech database demonstrate that the performance is improved.

This paper describes how speech recognition automatically uses the local language that originated from Indonesia, namely the Sundanese, which can control electronic devices in the home. Sundanese is the second most widely used regional language in Indonesia after Java. Voice recognition using Sundanese is carried out in this study to be able to control several electronic devices directly with high accuracy. The method used in the introduction of Sundanese is the Mel Frequency Cepstral Coefficient (MFCC) extraction method and identification of backpropagation-based neural networks. There are 10 Sundanese instructions used in the introduction to speech as input to the system, and each instruction has 2 to 3 Sundanese syllables. The output used by the author in this paper is five household electronic devices, to turn on or turn off controlled output using 2 Sundanese instructions. The data used in the backpropagation training process is 300 data; each instruction is 30 data. The results of training weights can produce an accuracy rate of 96.7% when testing the system. The control system uses the MFCC method and artificial neural network backpropagation algorithm so that the program works and Sundanese language voice recognition is good.

Speech recognition can be defined as the process of converting voice signals into the ranks of the word, by applying a specific algorithm that is implemented in a computer program. The research of speech recognition in Indonesia is relatively limited. This paper has studied methods of feature extraction which is the best among the Linear Predictive Coding (LPC) and Mel Frequency Cepstral Coefficients (MFCC) for speech recognition in Indonesian language. This is important because the method can produce a high accuracy for a particular language does not necessarily produce the same accuracy for other languages, considering every language has different characteristics. Thus this research hopefully can help further accelerate the use of automatic speech recognition for Indonesian language. There are two main processes in speech recognition, feature extraction and recognition. The method used for comparison feature extraction in this study is the LPC and MFCC, while the method of recognition using Hidden Markov Model (HMM). The test results showed that the LPC method is better than MFCC in Indonesian language speech recognition.

Speech production knowledge has been used to enhance the phonetic representation and the performance of automatic speech recognition (ASR) systems successfully. Representations of speech production make simple explanations for many phenomena observed in speech. These phenomena can not be easily analyzed from either acoustic signal or phonetic transcription alone. One of the most important aspects of speech production knowledge is the use of articulatory knowledge, which describes the smooth and continuous movements in the vocal tract. In this paper, we present a new articulatory model to provide available information for rescoring the speech recognition lattice hypothesis. The articulatory model consists of a feature front-end, which computes a voicing feature based on a spectral harmonics correlation (SHC) function, and a back-end based on the combination of deep neural networks (DNNs) and hidden Markov models (HMMs). The voicing features are incorporated with standard Mel frequency cepstral coefficients (MFCCs) using heteroscedastic linear discriminant analysis (HLDA) to compensate the speech recognition accuracy rates. Moreover, the advantages of two different models are taken into account by the algorithm, which retains deep learning properties of DNNs, while modeling the articulatory context powerfully through HMMs. Mandarin speech recognition experiments show the proposed method achieves significant improvements in speech recognition performance over the system using MFCCs alone.

Speech signal analysis aims to identify patterns within data to develop effec tive recognition algorithms. This process primarily utilizes feature extraction techniques such as linear predictive coding (LPC), linear predictive cepstral co efficients (LPCCs), and Mel-frequency cepstral coefficients (MFCCs). These features are crucial for constructing recognition algorithms that leverage both statistical and deep learning methods. While deep learning models require ex tensive datasets, they often prove unsuitable for low-resource languages. The Hidden Markov model (HMM)is the most widely adopted statistical framework in speech processing. However, HMMs are characterized by state-dependent models, where each state interacts only with its neighboring states. This limita tion restricts HMMs from capturing long-term signal properties, highlighting the need for addressing these constraints at the feature extraction stage. Most feature extraction methods rely on short-term signal processing, which further limits the comprehension of speech utterances. To overcome these limitations, alter native methods are necessary to capture more comprehensive patterns. This pa per presents a graph-based approach for analyzing speech trajectories and their temporal properties, which are subsequently validated using HMMs in speech recognition tasks. Graph-based representations on a low-resource Telugu dataset improve recognition accuracy by 13% while reducing processing time compared to traditional LPC.

Towards improving ASR robustness for PSN and GSM telephone applications

Improving mispronunciation detection using adaptive frequency scale

This paper aims to design and implement English digits speech recognition system using Matlab (GUI). This work was based on the Hidden Markov Model (HMM), which provides a highly reliable way for recognizing speech. The system is able to recognize the speech waveform by translating the speech waveform into a set of feature vectors using Mel Frequency Cepstral Coefficients (MFCC) technique This paper focuses on all English digits from (Zero through Nine), which is based on isolated words structure. Two modules were developed, namely the isolated words speech recognition and the continuous speech recognition. Both modules were tested in both clean and noisy environments and showed a successful recognition rates. In clean environment and isolated words speech recognition module, the multi-speaker mode achieved 99.5% whereas the speaker-independent mode achieved 79.5%. In clean environment and continuous speech recognition module, the multi-speaker mode achieved 72.5% whereas the speaker-independent mode achieved 56.25%. However in noisy environment and isolated words speech recognition module, the multi-speaker mode achieved 88% whereas the speaker-independent mode achieved 67%. In noisy environment and continuous speech recognition module, the multi-speaker mode achieved 82.5% whereas the speaker-independent mode achieved 76.67%. These recognition rates are relatively successful if compared to similar systems.

The more frequent human interaction to technology demands the development of methods of interaction with the machine to a more natural. Sound which is the most frequently used communication humans makes it one of the natural methods of interaction. Thus the development of a system that can recognize human speech as an action on the machine can be an option for those problems. Voice command is a speech recognition system to give the functions and actions on a system that has been defined as Command and Control systems. Amplitude value is taken from the input sound signal, so we get a set of real numbers is the value input for feature extraction. Feature extraction methods used in this paper are Mel frequency cepstral coefficient (MFCC). Early stages of MFCC is splitting the input signal amplitude values into frames which are processed using the mel-filterbak adapted from the human hearing system. The results of feature extraction made into a codebook which is used as an input symbol on HMM to form a model of every word. On testing step the characteristics of the test signal that has been quantized, then matched with the model that has been created in the previous step. From the test results, the system can recognize spoken words of speakers with an average accuracy of 93.89% in the environment without noise, and 58.1% in environments with noise.

Automatic speech recognition (ASR) system is used to recognize the text transcript from the given speech signal. Such speech signal can contain either isolated words or large vocabulary continuous speech (LVCS). Isolated words can be recognized with high accuracy in clean environment, but recognizing continuous words involves various parameters like speech corpus, speaker, environment noise, etc., that affects the accuracy of automatic speech recognition system directly. In the proposed work, hybrid feature extraction technique combines the perceptual linear predictive (PLP) and mel frequency cepstral coefficient (MFCC) to improve the accuracy of ASR in noisy environment. Voice activity and detection (VAD)-based frame dropping is used for improving the phonemes modeling by removing the pauses and distorted elements from the given speech signal. The proposed hybrid model with VAD is implemented by using self-generated speech corpus and shows relatively 12% increase in recognition rate compared with the state-of-the-art methodology.KeywordsAutomatic speech recognition (ASR)Perceptual linear predictive (PLP)Mel frequency cepstral coefficient (MFCC)Voice activity and detection (VAC)

A technique for suppressing the effect of any type of impulsive noise in the context of automatic speech recognition (ASR) is presented. The authors suggest the interposition in the mel frequency cepstral coefficients (MFCC) feature extraction front-end of (i) an identification stage responsible for the detection of corrupted cepstral vectors and (ii) a restoration stage that acts locally on the degraded cepstrum.