Speaker-independent Speech Recognition Research Articles

Isolated word recognition based on PNCC with different classifiers in a noisy environment

Speech Recognition is one of the important methods to control robotics, mobile applications, automobile, etc. Nevertheless, still, the Speech recognition rate needs to be improved especially in noisy environments. The main goal of this paper is to investigate the best features for isolated word recognition between the most common features extraction technique Mel Frequency Cepstral Coefficient MFCC and noise immunity features extraction technique Power Normalized Cepstral Coefficient PNCC, with the aid of four classifiers, Weighted KNN, Medium KNN, Linear SVM, Quadratic SVM in different levels of noise. The database that is used for this study is Speech Commands Data Set v0.01 which is containing 64,727 audio files to get a good estimation of the recognition rate. The algorithm based on MFCC is recommended for Speaker dependent speech recognition a little bit better (1.1%) than the algorithm based on PNCC in a noise-free environment. Nevertheless, the proposed algorithm based on PNCC is more efficient than the algorithm based on MFCC for Speaker independent speech recognition, and overall Recognition accuracy. Moreover, it has more immunity to noise and it is more accurate than MFCC by 8.7% on average with already trained speakers and 18.528% on average with new trained speakers’ tests. Finally, the proposed algorithm based on the PNCC feature extraction with aid of Weighted KNN is more accurate than other algorithms for Speaker independent speech recognition and overall Recognition accuracy. Moreover, it has more immunity to noise.

Domain Invariant Feature Learning for Speaker-Independent Speech Emotion Recognition

In this paper, we propose a novel domain invariant feature learning (DIFL) method to deal with speaker-independent speech emotion recognition (SER). The basic idea of DIFL is to learn the speaker-invariant emotion feature by eliminating domain shifts between the training and testing data caused by different speakers from the perspective of multi-source unsupervised domain adaptation (UDA). Specifically, we embed a hierarchical alignment layer with the strong-weak distribution alignment strategy into the feature extraction block to firstly reduce the discrepancy in feature distributions of speech samples across different speakers as much as possible. Furthermore, multiple discriminators in the discriminator block are utilized to confuse the speaker information of emotion features both inside the training data and between the training and testing data. Through them, a multi-domain invariant representation of emotional speech can be gradually and adaptively achieved by updating network parameters. We conduct extensive experiments on three public datasets, i. e., Emo-DB, eNTERFACE, and CASIA, to evaluate the SER performance of the proposed method, respectively. The experimental results show that the proposed method is superior to the state-of-the-art methods.

Voice Conversion Based Augmentation and a Hybrid CNN-LSTM Model for Improving Speaker-Independent Keyword Recognition on Limited Datasets

Keyword recognition is the basis of speech recognition, and its application is rapidly increasing in keyword spotting, robotics, and smart home surveillance. Because of these advanced applications, improving the accuracy of keyword recognition is crucial. In this paper, we proposed voice conversion (VC) - based augmentation to increase the limited training dataset and a fusion of a convolutional neural network (CNN) and long-short term memory (LSTM) model for robust speaker-independent isolated keyword recognition. Collecting and preparing a sufficient amount of voice data for speaker-independent speech recognition is a tedious and bulky task. To overcome this, we generated new raw voices from the original voices using an auxiliary classifier conditional variational autoencoder (ACVAE) method. In this study, the main intention of voice conversion is to obtain numerous and various human-like keywords’ voices that are not identical to the source and target speakers’ pronunciation. Parallel VC was used to accurately maintain the linguistic content. We examined the performance of the proposed voice conversion augmentation techniques using robust deep neural network algorithms. Original training data, excluding generated voice using other data augmentation and regularization techniques, were considered as the baseline. The results showed that incorporating voice conversion augmentation into the baseline augmentation techniques and applying the CNN-LSTM model improved the accuracy of isolated keyword recognition.

Speaker Adaptation Using i-Vector Based Clustering

We propose a novel speaker adaptation method using acoustic model clustering. The similarity of different speakers is defined by the cosine distance between their i-vectors (intermediate vectors), and various efficient clustering algorithms are applied to obtain a number of speaker subsets with different characteristics. The speaker-independent model is then retrained with the training data of the individual speaker subsets grouped by the clustering results, and an unknown speech is recognized by the retrained model of the closest cluster. The proposed method is applied to a large-scale speech recognition system implemented by a hybrid hidden Markov model and deep neural network framework. An experiment was conducted to evaluate the word error rates using Resource Management database. When the proposed speaker adaptation method using i-vector based clustering was applied, the performance, as compared to that of the conventional speaker-independent speech recognition model, was improved relatively by as much as 12.2% for the conventional fully neural network, and by as much as 10.5% for the bidirectional long short-term memory.

Hearing impaired speech recognition: Stockwell features and models

The development of speech recognition system for recognising the speeches of a reasonable person in various languages usually is in fashion and does not involve challenges to be faced by the researchers. For the past ten years, advances are taking place in analysing and recognising the speeches of the hearing impaired because of the deployment of sophisticated processing methods to study the characteristics of speech production. These technological advancements not only pave the way for developing algorithms for recognising the speeches of healthy persons, also recognising the utterances of the hearing impaired. The study on analysing the oral communication skills of hearing-impaired children has received the attention of the researchers, speech pathologists and audiologists to develop assistive tool/system because inadequacy of such skills dramatically affects the social, educational and career opportunities available to them at large. This paper mainly emphasises the need for the development of a more challenging speaker independent speech recognition system for hearing impaired so that the system can respond to the speech uttered by any HI. In this work, Modified Group Delay Features and Stockwell transform cepstral features are used at the front end and vector quantisation (VQ) and Multivariate Hidden Markov Models (MHMM) at the back end for recognising the speeches uttered by any person with hearing disability. Performance of the system is compared for the three modelling techniques VQ, Fuzzy C Means (FCM) clustering and MHMM for the recognition of isolated digits in Tamil. Recognition accuracy is 89.25% and 79.5% for speaker dependent and independent speech recognition system for the hearing impaired. Performance of the system reveals that this system may be deployed to understand the speeches uttered by any hearing impaired speaker, improve the social status of the people with hearing impairment and mitigate the social stigma in leading a normal life.

Mobile-Based Search Engine Dictionary for Information Technology Students

Modern technology uses speech recognition software to speed up the process of gathering inputs from the user to save significant amount of time for every process. While this process is complex in and of itself, it is further complicated by the fact (the user) from different parts of the country have varying accent and pronounce the same words differently. Therefore, the proponent intends to create a more speaker-independent Speech Recognition Search Engine for I.T. Dictionary to maintain the speed and accuracy of transcription. An increasing number of people use their smart phones for a several of reason either to connect with friends, conduct e-commerce transactions, browse the Internet, etc. But for those who in the I.T Field, there is an urgent need for a dictionary, to embrace this new mode of information dissemination. Due to a variety of reasons, people in Information Technology Students or even Professionals finds all I.T terms difficult to memorize or just even understand. They prefer to use their mobile smart phone to get information quickly to get the immediate answer for their query. This research study is an attempt to provide a mobile application solution.

A strategic approach to recognize the speech of the children with hearing impairment: different sets of features and models

The automatic speech recognition system is developed and tested for recognizing the speeches of a normal person in various languages. This paper mainly emphasizes the need for the development of a more challenging speaker independent speech recognition system for hearing impaired to recognize the speeches uttered by any Hearing Impaired (HI) speaker. In this work, Gamma tone energy features with filters spaced an equivalent rectangular bandwidth (ERB), MEL & BARK scale, and MFPLPC features are used at the front end and vector quantization (VQ) & multivariate hidden Markov models (MHMM) at the back end for recognizing the speeches uttered by any hearing impaired speaker. Performance of the system is compared for the three modeling techniques VQ, FCM (Fuzzy C means) clustering and MHMM for the recognition of isolated digits and simple continuous sentences in Tamil. Recognition accuracy (RA) is 81.5% with speeches of eight speakers considered for training and speeches of the remaining two speakers considered for testing for speaker independent isolated digit recognition system. Accuracy is found to be 91% and 87.5% for considering 90% of the data for training and 10% for testing for speaker independent isolated digit and continuous speech recognition systems respectively. Accuracy can be further enhanced by having an extensive database for creating models/templates. Receiver operating characteristics (ROC) drawn between True Positive Rate and False Positive Rate is used to assess the performance of the system for HI. This system can be utilized to understand the speech uttered by any hearing impaired speaker and the system facilitates the provision of necessary assistance to them. It ultimately improves the social status of the hearing impaired people and their confidence level will be enhanced.

The Echo™ as a Speaker-Independent Speech Recognition Device to Support Children with Autism: an Exploratory Study

Everyday, consumer-level technologies such as mobile technologies may lend themselves to be repurposed for special needs of the populations. In this three-part study, we aimed to investigate whether the Amazon Echo™, a popular personal assistant general consumer product, can function as a speaker-independent device that permits the hands-free retrieval of visual supports for children with autism. Phase 1 investigated whether the Echo’s speaker-independent speech recognition system, linked to a proprietary Amazon “Skill,” could retrieve visual supports in order to facilitate direction following (e.g., “put the girl under the bowl”). The accuracy with which the Echo retrieved and delivered visual supports to the iPad was found to be low, suggesting that the Echo cannot function as a speaker-independent speech recognition system. Subsequently, we customized the vocabulary that was delivered to the Echo with the relevant vocabulary and repeated the protocol in phase 2. A significant increase in accuracy was noted. Finally, in phase 3, the experimenter asked the Echo to retrieve visual supports in the presence of a child with autism while monitoring correct implementation of directives based on successful retrieval by the Echo. Results will be discussed in terms of implications for future research.

An experimental framework of speaker independent speech recognition system for Kashmiri language (K-ASR) system using Sphinx

Speech to text conversion in various languages have been performed so far but no process has definedfor the Kashmiri language. There has been no research done on Kashmiri speech recognition. So in thiswork, we describe the development as well as implementation of first CMU Sphinx-3 based speechrecognizer for the Kashmiri language. Recognition of the words have been done by using hidden markovmodels (HMMs). Dictionary consists of 100 words, representing Kashmiri digits from one (akh) tohundred (hat). Here, we developed a speaker independent, Kashmiri - Automatic Speech Recognition (KASR) system. The System is trained and tested for 1200 words spoken by 12 male and female speakers.Maximum Accuracy of 78.33% was achieved by the K-ASR system.

A speech-based distributed architecture platform for an intelligent ambience

In the paper, a speech-based platform for intelligent ambience and/or supportive environment applications is presented. The platform has a distributed architecture, which enables extended connectivity and support for multiple intelligent ambience services. The mobile unit Genesis is an integral part of the distributed platform, enabling interaction between several users and the environment. Furthermore, the sophisticated client/server platform's architecture incorporates robust speech recognition and text-to-speech synthesis engines for more natural human-machine interaction between users and the mobile unit Genesis. Both engines are multilingual oriented. Although the whole system is developed for the Slovenian language, it can be quickly adapted for other languages when appropriate language resources are available. With high speaker independent speech recognition accuracy and low command-to-operation delay, Genesis proves to have good manoeuvrability and it is easy to operate even by a non-experienced operator.

All I really need to know I learned from Lou and Nat: Lou Braida

Lou Braida was one of my two primary mentors in graduate school at MIT. Lou taught me innumerable things. In this talk, I will only have time to focus on a few items. I will describe how I applied to Speech Recognition what I learned from him about the power of a psychophysical approach to research problems, the importance of good data collection, and the value of long-term spectral characteristics in perception. Speech recognition by now is a relatively mature field, but at the time the field was relatively unexplored territory. Lou’s training allowed me to see ways to make advances in speech recognition experimental design, speaker-independent speech recognition, and noise-immune features and processing. While many of these early ideas have been subsumed over the years by more sophisticated processing, many of them have their roots in techniques inspired by a combination of perceptual knowledge with principled engineering design. Lou was, and is a master of both and I am forever grateful for his inspiration, mentoring, and friendship in shaping my career.

Speaker Independent Speech Recognition of Isolated Words in Room Environment

In this paper, the process of recognizing some important words from a large set of vocabularies is demonstrated based on the combination of dynamic and instantaneous features of the speech spectrum. There are many procedures to recognize a word by its vowel but this paper presents the highly effective speaker independent speech recognition in a typical room environment noise cases. To distinguish several isolated words of the sound of different vowels, two important features known as Pitch and Formant are extracted from the speech signals collected from a number of random male and female speakers. The extracted features are then analysed for the particular utterances to train the system. The specific objectives of this work are to implement an isolated and automatic word speech recognizer, which is capable of recognizing as well as responding to speech and an audio interfacing system between human and machine for an effective human-machine interaction. The whole system has been tested using computer codes and the result was satisfactory in almost 90% of cases. However, system might get confused by similar vowel sounds sometimes.

Croatian Large Vocabulary Automatic Speech Recognition

This paper presents procedures used for development of a Croatian large vocabulary automatic speech recognition system (LVASR). The proposed acoustic model is based on context-dependent triphone hidden Markov models and Croatian phonetic rules. Different acoustic and language models, developed using a large collection of Croatian speech, are discussed and compared. The paper proposes the best feature vectors and acoustic modeling procedures using which lowest word error rates for Croatian speech are achieved. In addition, Croatian language modeling procedures are evaluated and adopted for speaker independent spontaneous speech recognition. Presented experiments and results show that the proposed approach for automatic speech recognition using context-dependent acoustic modeling based on Croatian phonetic rules and a parameter tying procedure can be used for efficient Croatian large vocabulary speech recognition with word error rates below 5%.

Speaker Independent Speech Recognition Using Maximum Likelihood Approach for Isolated Words

Speaker Independent Speech Recognition Using Maximum Likelihood Approach for Isolated Words

Emotional speech feature normalization and recognition based on speaker-sensitive feature clustering

In this paper we propose a feature normalization method for speaker-independent speech emotion recognition. The performance of a speech emotion classifier largely depends on the training data, and a large number of unknown speakers may cause a great challenge. To address this problem, first, we extract and analyse 481 basic acoustic features. Second, we use principal component analysis and linear discriminant analysis jointly to construct the speaker-sensitive feature space. Third, we classify the emotional utterances into pseudo-speaker groups in the speaker-sensitive feature space by using fuzzy k-means clustering. Finally, we normalize the original basic acoustic features of each utterance based on its group information. To verify our normalization algorithm, we adopt a Gaussian mixture model based classifier for recognition test. The experimental results show that our normalization algorithm is effective on our locally collected database, as well as on the eNTERFACE'05 Audio-Visual Emotion Database. The emotional features achieved using our method are robust to the speaker change, and an improved recognition rate is observed.

Artificial Neural Network-Based Speech Recognition Using Dwt Analysis Applied On Isolated Words From Oriental Languages

Speech recognition is an emerging research area having its focus on human computer interactions (HCI) and expert systems. Analyzing speech signals are often tricky for processing, due to the non-stationary nature of audio signals. The work in this paper presents a system for speaker independent speech recognition, which is tested on isolated words from three oriental languages, i.e., Urdu,Persian, and Pashto. The proposed approach combines discrete wavelet transform (DWT) and feed-forward artificial neural network (FFANN) for the purpose of speech recognition. DWT is used for feature extraction and the FFANN is utilized for the classification purpose. The task of isolated word recognition is accomplished with speech signal capturing, creating a code bank of speech samples, and then by applying pre-processing techniques.For classifying a wave sample, four layered FFANN model is used with resilient back-propagation (Rprop). The proposed system yields high accuracy for two and five classes.For db-8 level-5 DWT filter 98.40%, 95.73%, and 95.20% accuracy rate is achieved with 10, 15, and 20 classes, respectively. Haar level-5 DWT filter shows 97.20%, 94.40%, and 91% accuracy ratefor 10, 15, and 20 classes, respectively. The proposed system is also compared with a baseline method where it shows better performance. The proposed system can be utilized as a communication interface to computing and mobile devices for low literacy regions.

Design of Voice Control System for Smart Home Based on STM32

The application of embedded speech recognition technology in the smart home is researched, combining of the Internet of Things, the voice control system for smart home has been designed. The core processor chooses the high-performance Cortex-M4 MCU STM32F407VGT6 produced by STMicroelectronics. The system contains a hardware unit based on LD3320 for speaker-independent speech recognition. RF wireless communication uses ultra-low power chip CC1101 and GSM employ SIM900A. Real-time operating system FreeRTOS is used for multitask scheduling and the operation of household devices. The practical application verifies that this voice control system practicably can identify voice commands quickly and accurately, complete the control actions primely, has a wide application prospect.

Speech Emotion Recognition Using Fourier Parameters

Recently, studies have been performed on harmony features for speech emotion recognition. It is found in our study that the first- and second-order differences of harmony features also play an important role in speech emotion recognition. Therefore, we propose a new Fourier parameter model using the perceptual content of voice quality and the first- and second-order differences for speaker-independent speech emotion recognition. Experimental results show that the proposed Fourier parameter (FP) features are effective in identifying various emotional states in speech signals. They improve the recognition rates over the methods using Mel frequency cepstral coefficient (MFCC) features by 16.2, 6.8 and 16.6 points on the German database (EMODB), Chinese language database (CASIA) and Chinese elderly emotion database (EESDB). In particular, when combining FP with MFCC, the recognition rates can be further improved on the aforementioned databases by 17.5, 10 and 10.5 points, respectively.

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.

Comparative Analysis of Adapted Foreign Language and Native Lithuanian Speech Recognizers for Voice User Interface

Paper presents research results obtained when building a speaker independent hybrid speech recognizer. This recognizer will be integrated as a phrase recognizer in a medical-pharmaceutical information system. The hybrid speech recognizer consists of two recognition components: an adapted commercial Microsoft Spanish speech recognizer and a locally developed hidden Markov models based recognizer implementing Lithuanian acoustic models. Efficiency of both recognition components was evaluated on multiple speaker independent speech recognition tasks. The average accuracy of Lithuanian recognizer was higher reaching 0.6% phrase error rate for user requests in medical-pharmaceutical domain. The adapted commercial Spanish speech recognizer showed the ability to improve the accuracy of Lithuanian recognizer in the worst recognition scenarios. These results proved the hypothesis formulated when proposing the basic idea of hybrid recognition approach: recognition errors from different recognizers built using various techniques are not strongly correlated. This fact could be exploited for improved overall speech recognition accuracy. DOI: http://dx.doi.org/10.5755/j01.eee.19.7.5171