Speaker-dependent 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.

Artificial intelligence and the dignity of risk

The increased use of AI-based systems poses risks and opportunities for people with cognitive disabilities. On the one hand, automated administrative systems, such as job applicant screeners, may disadvantage people whose patterns of strengths and weaknesses, and whose life circumstances, differ from those commonly seen in pools of data. On the other hand, people with cognitive disabilities stand to gain from AI's potential to provide superior support, such as speaker-dependent speech recognition. Further, privacy concerns are heightened, both because of greater likelihood that people with uncommon combinations of attributes can be identified from their data, and because of the potential for discrimination and exploitation. It is important that people with disabilities are able to make self-directed choices about the tradeoffs among risks and benefits, not denying them the dignity of risk that others have. Enabling this calls for advances both in technology and in organization.

Arabic isolated word recognition system using hybrid feature extraction techniques and neural network

In this paper, we implemented a speaker-dependent speech recognition system for 11 standard Arabic isolated words. During the feature extraction phase, several techniques were used such as Mel frequency cepstral coefficients, perceptual linear prediction, relative perceptual linear prediction and their first order temporal derivatives. Principal component analysis was adopted in order to reduce the feature dimension. The recognition phase is based on the feed forward back-propagation neural network using two learning algorithms: the Levenberg–Marquardt “Trainlm” and the scaled conjugate gradient “Trainscg”. Hybrid approaches were used and compared in terms of computational time and recognition rates and have produced very interesting performances.

Decoding Silent Speech in Japanese from Single Trial EEGS: Preliminary Results

We propose a new scheme for speaker-dependent silent speech recognition systems (SSRSs) using both single-trial electroencephalograms (EEGs) scalp-recorded and speech signals measured during overtly and covertly speaking “janken” and “season” in Japanese. This scheme consists of two phases. The learning phase specifies a Kalman filter using spectrograms of the speech signals and independent components (ICs), whose equivalent current dipole source localization (ECDL) solutions were located mainly at the Broca’s area, of the EEGs during the actual speech. In case of the “season” task, the speech signals were transformed into vowel and consonant sequences, and these relationships were learned by hidden Markov model (HMM) with Gaussian mixture densities. The decoding phase predicts spectrograms for the silent “janken” and “season” using the Kalman filter with the EEGs during the silent speech. For the silent “season”, the predicted spectrograms were inputted to the HMM, and which “season” was silently spoken was determined by the maximal log-likelihood among each HMM. Our preliminary results as training steps are as follows: the silent “jankens” were correctly discriminated; the silent “season”-HMMs worked well, suggesting that this scheme might be applied to the discrimination between all the pairs of the hiraganas

Voice Operated Micro Air Vehicle

This paper presents the design of a Micro Air Vehicle (MAV) operated by speech commands. A person from a base station can comfortably control the motion of the MAV by issuing speech commands. This system proposes to have two units, one in the base station which is the control unit and the other is a flying machine. The base station unit consists of a Digital Signal Processor (DSP) which performs a speaker dependent speech recognition algorithm for a predefined command library. The DSP generates control signals which are wirelessly transmitted to the MAV. A Field Programmable Gate Array (FPGA) is placed in the MAV for on-board operations. The FPGA receives the wirelessly transmitted control signals and alters the motion of the MAV based on the control signals received. The MAV consists of a single rotating wing with two rotors placed on either side of the fuselage, providing it with the capability of vertical take-off and landing (VTOL).

One-against-All Weighted Dynamic Time Warping for Language-Independent and Speaker-Dependent Speech Recognition in Adverse Conditions

Considering personal privacy and difficulty of obtaining training material for many seldom used English words and (often non-English) names, language-independent (LI) with lightweight speaker-dependent (SD) automatic speech recognition (ASR) is a promising option to solve the problem. The dynamic time warping (DTW) algorithm is the state-of-the-art algorithm for small foot-print SD ASR applications with limited storage space and small vocabulary, such as voice dialing on mobile devices, menu-driven recognition, and voice control on vehicles and robotics. Even though we have successfully developed two fast and accurate DTW variations for clean speech data, speech recognition for adverse conditions is still a big challenge. In order to improve recognition accuracy in noisy environment and bad recording conditions such as too high or low volume, we introduce a novel one-against-all weighted DTW (OAWDTW). This method defines a one-against-all index (OAI) for each time frame of training data and applies the OAIs to the core DTW process. Given two speech signals, OAWDTW tunes their final alignment score by using OAI in the DTW process. Our method achieves better accuracies than DTW and merge-weighted DTW (MWDTW), as 6.97% relative reduction of error rate (RRER) compared with DTW and 15.91% RRER compared with MWDTW are observed in our extensive experiments on one representative SD dataset of four speakers' recordings. To the best of our knowledge, OAWDTW approach is the first weighted DTW specially designed for speech data in adverse conditions.

A Novel Weighted Dynamic Time Warping for Light Weight Speaker-Dependent Speech Recognition in Noisy and Bad Recording Conditions

Lightweight speaker-dependent (SD) automatic speech recognition (ASR) is a promising solution for the problems of possibility of disclosing personal privacy and difficulty of obtaining training material for many seldom used English words and (often non-English) names. Dynamic time warping (DTW) algorithm is the state-of-the-art algorithm for small foot-print SD ASR applications, which have limited storage space and small vocabulary. In our previous work, we have successfully developed two fast and accurate DTW variations for clean speech data. However, speech recognition in adverse conditions is still a big challenge. In order to improve recognition accuracy in noisy and bad recording conditions, such as too high or low recording volume, we introduce a novel weighted DTW method. This method defines a feature index for each time frame of training data, and then applies it to the core DTW process to tune the final alignment score. With extensive experiments on one representative SD dataset of three speakers' recordings, our method achieves better accuracy than DTW, where 0.5% relative reduction of error rate (RRER) on clean speech data and 7.5% RRER on noisy and bad recording speech data. To the best of our knowledge, our new weighted DTW is the first weighted DTW method specially designed for speech data in noisy and bad recording conditions.

남녀성별 분류를 위한 화자종속 음성인식 알고리즘

This paper proposes a speaker-dependent speech recognition algorithm which can classify the gender for male and female speakers in white noise and car noise, using a neural network. The proposed speech recognition algorithm is trained by the neural network to recognize the gender for male and female speakers, using LPC (Linear Predictive Coding) cepstrum coefficients. In the experiment results, the maximal improvement of total speech recognition rate is 96% for white noise and 88% for car noise, respectively, after trained a total of six neural networks. Finally, the proposed speech recognition algorithm is compared with the results of a conventional speech recognition algorithm in the background noisy environment.

Improvement of the Linear Predictive Coding with Windowed Autocorrelation

In this paper, we propose a new procedure for improvement of the linear predictive coding. To reduce the error power incurred by the coding, we interchanged the order of the two procedures of windowing on the signal and linear prediction. This scheme corresponds to LPC extraction with windowed autocorrelation. The proposed method requires more calculational time because it necessitates matrix inversion on more parameters than the conventional technique where an efficient Levinson-Durbin recursive procedure is applicable with smaller parameters. Experimental test over various speech phonemes showed, however, that our procedure yields about 5 % less power distortion compared to the conventional technique. Consequently, the proposed method in this paper is thought to be preferable to the conventional technique as far as the fidelity is concerned. In a separate study of speaker-dependent speech recognition test for 50 isolated words pronounced by 40 people, our approach yielded better performance too.

A speech-controlled environmental control system for people with severe dysarthria

Automatic speech recognition (ASR) can provide a rapid means of controlling electronic assistive technology. Off-the-shelf ASR systems function poorly for users with severe dysarthria because of the increased variability of their articulations. We have developed a limited vocabulary speaker dependent speech recognition application which has greater tolerance to variability of speech, coupled with a computerised training package which assists dysarthric speakers to improve the consistency of their vocalisations and provides more data for recogniser training. These applications, and their implementation as the interface for a speech-controlled environmental control system (ECS), are described. The results of field trials to evaluate the training program and the speech-controlled ECS are presented. The user-training phase increased the recognition rate from 88.5% to 95.4% ( p < 0.001). Recognition rates were good for people with even the most severe dysarthria in everyday usage in the home (mean word recognition rate 86.9%). Speech-controlled ECS were less accurate (mean task completion accuracy 78.6% versus 94.8%) but were faster to use than switch-scanning systems, even taking into account the need to repeat unsuccessful operations (mean task completion time 7.7 s versus 16.9 s, p < 0.001). It is concluded that a speech-controlled ECS is a viable alternative to switch-scanning systems for some people with severe dysarthria and would lead, in many cases, to more efficient control of the home.

Speaker adaptation techniques for speech recognition using probabilistic models

AbstractIn speech recognition, speaker adaptation refers to the range of techniques whereby a speech recognition system is adapted to the acoustic features of a specific user using a small sample of utterances from that user. In recent years the practical development of speaker‐independent speech recognition systems using continuous density hidden Markov models has seen significant progress; however, the recognition performance of these systems has not yet reached that of speaker‐dependent speech recognition systems in which a user's speech is registered beforehand. Much hope has therefore been placed on the establishment of speaker adaptation techniques that can bring performance of a speaker‐independent system up to that of a speaker‐dependent one using the smallest amounts of data. In this paper we present a survey of previous research into speaker adaptation techniques focusing particularly on three important approaches in this area: maximum a posteriori (MAP) parameter estimation, maximum likelihood linear regression (MLLR), and eigenvoices. We also discuss approaches that combine these techniques in a lateral fashion. © 2005 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 88(12): 25–42, 2005; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ ecjc.20207

Simultaneous recognition of words and prosody in the Boston University Radio Speech Corpus

This paper describes automatic speech recognition systems that satisfy two technological objectives. First, we seek to improve the automatic labeling of prosody, in order to aid future research in automatic speech understanding. Second, we seek to apply statistical speech recognition models of prosody for the purpose of reducing the word error rate of an automatic speech recognizer. The systems described in this paper are variants of a core dynamic Bayesian network model, in which the key hidden variables are the word, the prosodic tag sequence, and the prosody-dependent allophones. Statistical models of the interaction among words and prosodic tags are trained using the Boston University Radio Speech Corpus, a database annotated using the tones and break indices (ToBI) prosodic annotation system. This paper presents both theoretical and empirical results in support of the conclusion that a prosody-dependent speech recognizer—a recognizer that simultaneously computes the most-probable word labels and prosodic tags—can provide lower word recognition error rates than a standard prosody-independent speech recognizer in a multi-speaker speaker-dependent speech recognition task on radio speech.

Isolated word recognition using neural networks

This paper presents the speech recognition system, designed for recognition of 50 isolated Lithuanian words. The system is based on feedforward neural networks trained by error backpropagation algorithm. Speaker dependent speech recognition was investigated. The article presents analysis of the developed system parameters influence on the accuracy of word recognition. Experimental results showed that proposed system could be used for isolated word recognition.

A Mixture of Recurrent Neural Networks for Speaker Normalisation

In spite of recent advances in automatic speech recognition, the performance of state-of-the-art speech recognisers fluctuates depending on the speaker. Speaker normalisation aims at the reduction of differences between the acoustic space of a new speaker and the training acoustic space of a given speech recogniser, improving performance. Normalisation is based on an acoustic feature transformation, to be estimated from a small amount of speech signal. This paper introduces a mixture of recurrent neural networks as an effective regression technique to approach the problem. A suitable Viterbi-based time alignment procedure is proposed for generating the adaptation set. The mixture is compared with linear regression and single-model connectionist approaches. Speaker-dependent and speaker-independent continuous speech recognition experiments with a large vocabulary, using Hidden Markov Models, are presented. Results show that the mixture improves recognition performance, yielding a 21% relative reduction of the word error rate, i.e. comparable with that obtained with model-adaptation approaches.

Single-chip speech recognition system based on 8051 microcontroller core

This paper describes a single-chip speech recognition system. It contains the speech functions of prompt, playback, speaker-dependent speech recognition, suitable for the voice activated systems in toys, games, consumer electronics, office devices, etc. The chip is designed based on the SOC (system on chip) philosophy and an 8-bit MCU, RAM, ROM, ADC/DAC, PWM, I/O ports and other peripheral circuits are all embedded in it. Software modules including control/communication, speech coding and speech recognition algorithms are implemented in an 8051 compatible microcontroller core, resulting in the extremely low cost of the chip. The speech recognition adopts the template matching technique. It recognizes up to 20 phrases with an average length of 1 second and the recognition accuracy reaches more than 95% with the background SNR above 10 dB. Speech coding uses continuous variable slope delta modulation (CVSD) algorithm. The bit rate is 16 kbits/s.

Methods and apparatus for generating and using speaker independent garbage models for speaker dependent speech recognition purpose

Methods and apparatus for generating and using both speaker dependent and speaker independent garbage models in speaker dependent speech recognition applications are described. The present invention recognizes that in some speech recognition systems, e.g., systems where multiple speech recognition operations are performed on the same signal, it may be desirable to recognize and treat words or phrases in one part of the speech recognition system as garbage or out of vocabulary utterances with the understanding that the very same words or phrases will be recognized and treated as in-vocabulary by another portion of the system. In accordance with the present invention, in systems where both speaker independent and speaker dependent speech recognition operations are performed independently, e.g., in parallel, one or more speaker independent models of words or phrases which are to be recognized by the speaker independent speech recognizer are included as garbage (OOV) models in the speaker dependent speech recognizer. This reduces the risk of obtaining conflicting speech recognition results from the speaker independent and speaker dependent speech recognition circuits. The present invention also provides for the generation of speaker dependent garbage models from the very same data used to generate speaker dependent speech recognition models, e.g., word models. The technique involves processing the data included in the speaker dependent speech recognition models to create one or more speaker dependent garbage models.

Improving Speech Recognition Accuracy for Small Vocabulary Applications in Adverse Environments

The problem of improving the accuracy of small vocabulary isolated word speaker dependent speech recognition under adverse conditions such as factory environments is considered. A new approach to solving this problem, by using Output Probability Distributions (OPDs), is presented. OPDs improve the system performance by modelling inter-word relationships, something that a standard maximum likelihood (ML) technique fails to do. The system was tested using the TI46 database, corrupted with the NOISEX-92 database, as well as in a real-world factory environment, and achieved good results.

Speaker-dependent speech recognition using speaker independent models

The memory and data management requirements for text independent speaker dependent recognition are drastically reduced by using a novel approach that eliminates the need for separate acoustic recognition models for each speaker. This is achieved by using speaker independent recognition models at the acoustic level. The speaker dependent data stored for each item to be recognized consists only of information needed to determine the speaker independent recognition model sequence for that item.

Projection-based group delay scheme for speech recognition

This paper deals with speech recognition in a noisy environment. Some robust properties of the group delay spectrum are derived, and a projection-based group delay scheme is proposed for noisy speech recognition. Experiments show that the proposed new scheme obtains better recognition results, especially under low signal-to-noise ratio (SNR) conditions. When the features of speech signals are obtained from the group delay spectrum, the continuous hidden Markov models (CHMMs) are used for speaker-dependent speech recognition. The robustness of the noisy group delay spectrum is analyzed through a noisy speech model and histograms, respectively. We apply the projection measure to the CHMM and derive the correlative formulas.

The metamorphic algorithm: a speaker mapping approach to data augmentation

Large vocabulary speaker-dependent speech recognition systems adjust to the acoustic peculiarities of each new speaker based on some enrolment data provided by this speaker. As the amount of data required increases with the sophistication of the underlying acoustic models, the enrolment may get lengthy. To streamline it, it is therefore desirable to make use of previously acquired speech data. The authors describe a data augmentation strategy based on a piecewise linear mapping between the feature space of a new speaker and that of a reference speaker. This speaker-normalizing mapping is used to transform the previously acquired data of the reference speaker onto the space of the new speaker. The performance of the resulting procedure, dubbed the metamorphic algorithm, is illustrated on an isolated utterance speech recognition task with a vocabulary of 20000 words. Results show that the metamorphic algorithm can substantially reduce the word error rate when only a limited amount of enrolment data is available. Alternatively, it leads to a level of performance comparable to that obtained when a much greater amount of enrolment data is required from the new speaker. In addition, it can also be used for tracking spectral evolution over time, thus providing a possible means for robust speaker self-adaptation. >