Speaker-independent Recognition System Research Articles

A Proposal of Low Cost Home Automation System Using IoT and Voice Recognition

Home Automation System is becoming more popular day by day due to its numerous benefits. This project proposes an idea in the design of low cost home automation system by using the Internet of Things (IoT) and voice recognition. The layout of the home divided into four areas and each area has own function and system. The Raspberry Pi 3 (RPi) Model B+ used as the main controller for the processing and transmitting the input data. IoT provided huge storage for data collection from sensors and home appliances. An Android application is developed to monitor the home environment and remotely control the home devices by using the button or voice. The speaker-independent recognition system by using Google Voice to Text on Android embedded in this project for physically challenged people to control the electrical appliances without moving. All the data will be stored in Firebase and can be retrieved at any time by the application and the RPi board. There is a side view of a prototype model with two floors and divided into four home areas. This Low-Cost Home Automation System using IoT and Voice Recognition is successfully achieved the project’s objective.

Study of MFCC and IHC Feature Extraction Methods With Probabilistic Acoustic Models for Speaker Biometric Applications

Voice is an important human trait in natural human-to-human interaction / communication for identifying a person. So voice can be regarded as a biometric measure for recognizing or identifying the person similar to other biometric measures such as face, iris and fingerprints. Speaker recognition is a class of voice recognition where speaker is identified from the speech rather than the message. Automatic speaker recognition (SR) is an approach to identify people based on features extracted from speech utterances. The major task in any speaker recognition is to extract useful features and allow meaningful patterns of speaker models. This paper compares the performance of two feature extraction techniques Mel Frequency Cepstral Coefficient (MFCC) and Inner Hair Cell Coefficient (IHC) with two different modelling methods Gaussian Mixture Model - Universal background model (GMM - UBM) and i- vector approach. In this experiment speech samples of 600 speakers from TIMIT database with 10 utterances of each speaker are taken for identifying the speaker. A text independent speaker recognition system was implemented and this study resulted in an inference of MFCC feature outperforms IHC feature for both GMM and i vector. The performance of voiced speech IHC feature which simulates the physiological behaviour of human ear is better in terms of accuracy than full speech (voiced and unvoiced) in GMM.

An automatic speech recognition system for spontaneous Punjabi speech corpus

Automatic speech recognition is the central part of the wheel towards the natural person-to-machine interaction technique. Due to the high disparity of speaking styles, speech recognition surely demands composite methods to constitute this irregularity. A speech recognition method can work in numerous distinct states such as speaker dependent/independent speech, isolated/continuous/spontaneous speech recognition, for less to very large vocabulary. The Punjabi language is being spoken by concerning 104 million peoples in India, Pakistan and other countries with Punjabi migrants. The Punjabi language is written in Gurmukhi writing in Indian Punjab, while in Shahmukhi writing in Pakistani Punjab. In the paper, the objective is to build the speaker independent automatic spontaneous speech recognition system for the Punjabi language. The system is also capable to recognize the spontaneous Punjabi live speech. So far, no work has to be achieved in the area of spontaneous speech recognition system for the Punjabi language. The user interfaces for Punjabi live speech system is created by using the java programming. Till now, automatic speech system is trained with 6012 Punjabi words and 1433 Punjabi sentences. The performance measured in terms of recognition accuracy which is 93.79% for Punjabi words and 90.8% for Punjabi sentences.

Comparison of Text Independent Speaker Identification Systems using GMM and i-Vector Methods

Voice biometric provides a person’s unique identity and hence is widely employed for security applications. Here, Text Independent Speaker Recognition system is implemented using Gaussian Mixture Models (GMM) and i-Vector method with two features PNCC (Power Normalized Cepstral Coefficients) and RASTA PLP (Relative Spectral Perceptual Linear Prediction) coefficients. It was observed that accuracy for speaker identification is better when pitch and formants are appended to basic features. Also, the accuracy of i-vector method with PLDA (Probabilistic Linear Discriminant Analysis) classifier is better than that with CDS (Cosine Distance Scoring) classifier. Further, the performance improves when longer utterances are used.

Integrated models and features-based speaker independent emotion recognition

Speech emotion recognition has become a challenging task in speech technology in order to ensure better and effective human-machine interaction. Improving the accuracy of speaker independent emotion recognition system by using a database containing the speech samples of limited number of utterances spoken by limited set of speakers is more challenging. This paper mainly discusses the use of integrated features and integrated models for improving the accuracy of the emotion recognition system. Integrated features are obtained by concatenating the probability with the normal system features. Integrated models are obtained by combining clustering and continuous density hidden Markov models. Integrated models with Mel frequency perceptual linear predictive cepstrum concatenated with probability provides better accuracy of 89% for recognising emotions using EMO-DB which contains speech samples of ten different utterances of ten different speakers in different emotions such as anger, boredom, disgust, fear, happy, neutral and sad by applying spectral analysis as a additional pre processing technique.

Text Independent Automatic Speaker Recognition System using fusion of features

Text Independent Automatic Speaker Recognition System using fusion of features

IMPLEMENTATION OF HIERARCHICAL PHONEME CLASSIFICATION APPROACH ON LTDIGITS CORPORA

Better discrimination of phonemic units still remains one of the most important problems in automatic speech recognition. Direct phoneme recognition in speaker independent automatic speech recognition systems is unable to provide good enough recognition results. There is made an assumption that better results could be achieved through the recognition of phoneme groups using group characteristic features: voiced/unvoiced, vowel/consonant, etc. This paper describes and statistically motivates features and rules for the detection of phoneme groups using phonetically labeled data. Algorithms for recognition of stop and fricative consonants are presented. Experimental research confirmed the advantages of the hierarchical classification of phonemes. Combination of knowledge and rules for detection of acoustic events with the classical statistical classification methods produced an overall 3% improvement of phoneme recognition accuracy and a 52-55% reduction of time taken by classification.

ANALYSIS OF VOCAL PHONEMES AND FRICATIVE CONSONANT DISCRIMINATION BASED ON PHONETIC ACOUSTICS FEATURES

Direct recognition of phonemes in speaker independent recognition systems still cannot guarantee good enough recognition results. Here we want to investigate assumption that recognition could be improved via group features of phonemic system. We propose to try to recognize group of phoneme first (voiced/unvoiced, vowel/ con-sonant, etc.) then try to recognize phoneme itself. In this experiment a system for discrimination of fricative consonants and sonants was developed using phonetic – acoustic features.

Study of feature combination using HMM and SVM for multilingual Odiya speech emotion recognition

Emotions are broad aspects and expressed in a similar way by every human being; however, these are affected by culture. This creates a major threat to the universality of speech emotion detection system. Cultural behaviour of society affects the way emotions are expressed and perceived. Hence, an emotion recognition system customized for languages within a cultural group is feasible. In this work, a speaker dependent and speaker independent emotion recognition system has been proposed for two different dialects of Odisha: Sambalpuri and Cuttacki. Spectral speech features, such as, log power, Mel-frequency cepstral coefficients (MFCC), Delta MFCC, Double delta MFCC, log frequency power coefficients, and linear predictive cepstral coefficients, are used with Hidden Markov model and support vector machines (SVM) classifier, for classifying a speech into one of the seven discrete emotion classes: anger, happiness, disgust, fear, sadness, surprise, and neutral. For a better comparative study of system’s accuracy, features are taken individually as well as in combinations by varying sampling frequency, frame length and frame overlapping. Best average recognition accuracy obtained for speaker independent system, is 82.14 % for SVM classifier using only MFCC as feature vector. However, for speaker dependent system a hike in accuracy of more than 10 % is seen. It is also revealed that use of MFCC on SVM classifier, not only gives the best overall performance on 8 kHz sampling frequency, but also shows consistent performance for all the emotion classes, compared to other classifiers and feature combinations with less computational complexity. Hence, it can be applied efficiently in call centre application for emotion recognition over telephone.

Speaker Independent Speech Recognition using MFCC with Cubic-Log Compression and VQ Analysis

Speech processing is developed as one of the paramount requisition region of digital signal processing. Different fields for research in speech processing are speech recognition, speaker identification, speech bland, speech coding etc. The objective of Speaker Independent Speech Recognition is to concentrate, describe and distinguish information about speech signal and methodology towards creating the speaker free speech recognition system. Extracted information will be valuable for the directing and working different electronic contraptions and hardware through the human voice proficiently. Feature extraction is the first venture for speech recognition. Numerous algorithms are recommended / created by the scientists for feature extraction. In this work, the cubiclog compression in Mel-Frequency Cepstrum Coefficient (MFCC) feature extraction system is utilized to concentrate the characteristics from speech sign for outlining a speaker independent speaker recognition system. Extracted features are used to train and test this system with the help of Vector Quantization approach.

Speech Recognition using Wavelets and Improved SVM

Speaker recognition (identification/verification) is the computing task of validating a user’s claimed identity using speaker specific information included in speech waves: that is, it enables access control of various services by voice. Discrete Wavelet Transform (DWT) based systems for speaker recognition have shown robust results for several years and are widely used in speaker recognition applications. This paper is based on text independent speaker recognition system that makes use of Discrete Wavelet Transform (DWT) as a feature extraction and kernel Support Vector Machine (SVM) approach as a classification tool for taking the decision through applying simplified-Class Support Vector Machine approach. The proposed SVM approach can convert local Euclidean distances between frame vectors to angles by projecting these -dimensional vectors together, and get the minimum global distance from the non-linear aligned speech path in order to address audio classification, and hence, sound recognition.The DWT for each frame of the spoken word are taken as a tool for extracting the main feature as a data code vectors, next these data is normalized utilizing the normalized power algorithm that is used to reduce the number of feature vector coefficients then these data is scaled and tested with those stored of the training spoken words to achieve the speaker identification tasks, also the DWT gives fixed amount of data that can be utilized modesty by SVM.Finally, the proposed method is tested and trained upon a very large data base with results limited to ten speakers only (5 males and 5 females) with words of maximally 17 phenomena and its performance gives an accurate and stable results which rises the algorithm efficiency and reduce the execution time with 97% overall accuracy.

Rapid speaker adaptation in latent speaker space with non-negative matrix factorization

A novel speaker adaptation algorithm based on Gaussian mixture weight adaptation is described. A small number of latent speaker vectors are estimated with non-negative matrix factorization (NMF). These latent vectors encode the distinctive systematic patterns of Gaussian usage observed when modeling the individual speakers that make up the training data. Expressing the speaker dependent Gaussian mixture weights as a linear combination of a small number of latent vectors reduces the number of parameters that must be estimated from the enrollment data. The resulting fast adaptation algorithm, using 3s of enrollment data only, achieves similar performance as fMLLR adapting on 100+s of data. In order to learn richer Gaussian usage patterns from the training data, the NMF-based weight adaptation is combined with vocal tract length normalization (VTLN) and speaker adaptive training (SAT), or with a simple Gaussian exponentiation scheme that lowers the dynamic range of the Gaussian likelihoods. Evaluation on the Wall Street Journal tasks shows a 5% relative word error rate (WER) reduction over the speaker independent recognition system which already incorporates VTLN. The WER can be lowered further by combining weight adaptation with Gaussian mean adaptation by means of eigenvoice speaker adaptation.

A Comparitive Survey of ANN and Hybrid HMM/ANN Architectures for Robust Speech Recognition

This paper proposes two hybrid connectionist structural acoustical models for robust context independent phone like and word like units for speaker-independent recognition system. Such structure combines strength of Hidden Markov Models (HMM) in modeling stochastic sequences and the non-linear classification capability of Artificial Neural Networks (ANN). Two kinds of Neural Networks (NN) are investigated: Multilayer Perceptron (MLP) and Elman Recurrent Neural Networks (RNN). The hybrid connectionist-HMM systems use discriminatively trained NN to estimate the a posteriori probability distribution among subword units given the acoustic observations. We efficiently tested the performance of the conceived systems using the TIMIT database in clean and noisy environments with two perceptually motivated features: MFCC and PLP. Finally, the robustness of the systems is evaluated by using a new preprocessing stage for denoising based on wavelet transform. A significant improvement in performance is obtained with the proposed method.

Improved wavelet feature extraction using kernel analysis for text independent speaker recognition

A text independent speaker recognition system based on improved wavelet transform is proposed. Learning of the correlation between the wavelet transform and the expression vector is performed by kernel canonical correlation analysis. Kernel canonical correlation analysis is a nonlinear extension of canonical correlation analysis. Moreover, we also propose an improved kernel canonical correlation algorithm to tackle the singularity problem of the wavelet matrix. The identification model underlying the Gaussian mixture model is presented; in particular, an expectation-maximization algorithm is also proposed for adjusting the parameters. The experimental results on the TALUNG database and KING database illustrate the effectiveness of the proposed method.

Information enquiry kiosk with multimodal user interface

A multimodal interactive dialogue automaton (kiosk) for self-service is presented in the paper. Multimodal user interface allow people to interact with the kiosk by natural speech, gestures additionally to the standard input and output devices. Architecture of the kiosk contains key modules of speech processing and computer vision. An array of four microphones is applied for far-field capturing and recording of user’s speech commands, it allows the kiosk to detect voice activity, to localize sources of desired speech signals, and to eliminate environmental acoustical noises. A noise robust speaker-independent recognition system is applied to automatic interpretation and understanding of continuous Russian speech. The distant speech recognizer uses grammar of voice queries as well as garbage and silence models to improve recognition accuracy. Pair of portable video-cameras are applied for vision-based detection and tracking of user’s head and body position inside of the working area. Russian-speaking talking head serves both for bimodal audio-visual speech synthesis and for improvement of communication intelligibility by turning the head to an approaching client. Dialogue manager controls the flow of dialogue and synchronizes sub-modules for input modalities fusion and output modalities fission. The experiments made with the multimodal kiosk were directed to cognitive and usability studies of human-computer interaction by different communication means

Multi-segment vector quantization codebooks-based speaker independent word speech recognition system with high recognition rate.

Multi-segment vector quantization codebooks-based speaker independent word speech recognition system with high recognition rate.

Further reduced form of wavelet feature for text independent speaker recognition

A text independent speaker recognition system based on wavelet transform derived from fuzzy c-means clustering is proposed. The fuzzy c-means clustering is applied to the speaker data compression in spectrum domain. A set of experiments are conducted, which gives a 95% recognition rate for 100 Mandarin speakers.

Further reduced form of wavelet feature for text independent speaker recognition

A text independent speaker recognition system based on wavelet transform derived from fuzzy c-means clustering is proposed. The fuzzy c-means clustering is applied to the speaker data compression in spectrum domain. A set of experiments are conducted, which gives a 95% recognition rate for 100 Mandarin speakers.

Limited‐Vocabulary Estonian Continuous Speech Recognition System using Hidden Markov Models

The article presents a limited-vocabulary speaker independent continuous Estonian speech recognition system based on hidden Markov models. The system is trained using an annotated Estonian speech database of 60 speakers, approximately 4 hours in duration. Words are modelled using clustered triphones with multiple Gaussian mixture components. The system is evaluated using a number recognition task and a simple medium-vocabulary recognition task. The system performance is explored by employing acoustic models of increasing complexity. The number recognizer achieves an accuracy of 97%. The medium-vocabulary system recognizes 82.9% words correctly if operating in real time. The correctness increases to 90.6% if real-time requirement is discarded.

A rejection model based on multi-layer perceptrons for Mandarin digit recognition

High performance Mandarin digit recognition (MDR) is much more difficult to achieve than its English counterpart, especially on inexpensive hardware implementation. In this paper, a new Multi-Layer Perceptrons (MLP) based postprocessor, an a posteriori probability estimator, is presented and used for the rejection model of the speaker independent Mandarin digit recognition system based on hidden Markov model (HMM). Poor utterances, which are recognized by HMMs but have low a posteriori probability, will be rejected. After rejecting about 4.9% of the tested utterances, the MLP rejection model can boost the digit recognition accuracy from 97.1% to 99.6%. The performance is better than those rejection models based on linear discrimination, likelihood ratio or anti-digit.