Search

The performance of a speech recognizer depends upon the quality and robustness of the acoustic and features extracted from the speech wave. Most of the speech parameterization methods proposed so far are not very robust against variations in the speech signal. In this paper we will go through some of the various approaches used so far.

This paper investigates the use of a Time Delay Neural Network (TDNN) as fuzzy vector quantizer to improve the Distributed scheme of HMM speech recognition. We investigate how to optimize the use of the Vector Quantization (VQ) by combining complementary preprocessing techniques based on multi-streams acoustic analysis. Then, in order to eliminate the effect of quantization error incurred by the vector quantizer front-end process a Distributed TDNN Fuzzy Vector Quantizer (DTDNN-FVQ) scheme is proposed. The evaluation of the whole of these methods is performed by focusing on specific Arabic phonemes: emphatic and back consonants. Experimental results shows that the distributed approach proposed increases the global performance of the HMM speech recognition system.

The paper proposes a human-computation-based scheme for transcribing Southern Min speech corpora. The core idea is to implement a Web-based language learning system to collect orthographic and phonetic labels from a large amount of language learners and choose the commonly input labels as the transcriptions of the corpora. It is essentially a technology of distributed knowledge acquisition. Some computer-aided mechanisms are also used to verify the collected transcriptions. The benefit of the scheme is that it makes the transcribing task neither tedious nor costly. No significant budget should be made for transcribing large corpora. The design of a system for transcribing Min Nan speech corpora is described in detail. The application of a prototype version of the system shows that this transcribing scheme is an effective and economical way to generate orthographic and phonetic transcriptions.

ISBN : 978-1-4244-0778-1; International audience; This paper presents several new confidence measures for speech recognition applications. The major advantage of these measures is that they can be evaluated with only a part of the whole sentence. Two of these measures can be computed directly within the first step of the recognition process, synchronously with the decoding engine. Such measures are useful to drive the recognition process by modifying the likelihood score or to validate recognized words in on-the-fly applications as keyword spotting task and on-line automatic speech transcription for deaf people. Two kinds of results are given. Firstly, an EER evaluation on a French broadcast news corpus shows performance close to the batch version of these measures (23.9% against 23.8% of EER). Secondly, for the keyword spotting application, our best measure provides a decrease of the false-acceptation rate by 50% with only a decrease of the correct words by 5%.

In this paper we present an automated method for the classification of the origin of non-native speakers. The origin of non-native speakers could be identified by a human listener based on the detection of typical pronunciations for each nationality. Thus we suppose the existence of several phoneme sequences that might allow the classification of the origin of non-native speakers. Our new method is based on the extraction of discriminative sequences of phonemes from a non-native English speech database. These sequences are used to construct a probabilistic classifier for the speakerspsila origin. The existence of discriminative phone sequences in non-native speech is a significant result of this work. The system that we have developed achieved a significant correct classification rate of 96.3% and a significant error reduction compared to some other tested techniques.

Colloque avec actes et comité de lecture. internationale.; International audience; Support Vector Machines (SVM) represent a new approach to pattern classification developed from the theory of structural risk minimisation. In this paper, we propose an investigation into the application of SVM to the confidence measurement problem in speech recognition. Confidence measures are computed using the phone level information provided by a Hidden Markov Model (HMM) based speech recognizer. We use three kinds of average techniques as arithmetic, geometric and harmonic averages in order to compute a confidence measure for each word. The acceptance/rejection decision for a given word is based on the confidence feature vector which is processed by a SVM classifier. The performance of the proposed SVM classifier is compared with methods based on the averaging of phone confidence measures.

Support vector machines (SVM) is a new and very promising classification technique developed from the theory of structural risk minimisation. We propose an alternative out-of-vocabulary word detection method relying on confidence measures and support vector machines. Confidence measures are computed from phone level information provided by a hidden Markov model (HMM) based speech recognizer. We use three kinds of average techniques as arithmetic, geometric and harmonic averages to compute a confidence measure for each word. The acceptance/rejection decision of a word is based on the confidence feature vector which is processed by a SVM classifier. The performance of the proposed SVM classifier is compared with methods based on the averaging of confidence measures.

The authors propose a novel approach to neuronlike models that consists in simulating cortical columns, i.e. associations of neurons having a specific, functional activity. This model is built according to a theory which is consistent with neurobiological data. The model is implemented through a network of columns that simulate the inherently parallel functioning of the nervous system. A description is given of this model, and the experimental conditions in which it is now being tested are described. >

The authors propose a homogeneous architecture for organizing and controlling multiple knowledge sources in signal interpretation systems based on the decomposition of the interpretation problem into subproblems at successive conceptual levels. Partial interpretations and knowledge about the problem are partitioned into associations each of which consisting of several specialists. Information exchange is assured by a common data structure within knowledge sources at each given level and by a message passing mechanism between two different levels. Different strategies adapted to the nature of the problem may be implemented at each level. The interpretation process consists in executing the groups of knowledge sources in multiple phases controlled by data and models in each concept domain and allowing incremental construction of solutions. The authors illustrate the architecture by an application to continuous spoken Chinese understanding. >

The expectation-maximization (EM) algorithm is a general technique for maximum likelihood estimation (MLE). In this paper, we present two important theoretical issues concerning Gaussian mixture modeling (GMM) within the EM framework. First, we propose an EM algorithm for estimating the parameters of a GMM structure dedicated to speaker recognition, the probabilistic mapping network (PMN), where the Gaussian probability density function is realized as an internal node. Hence, the EM algorithm is extended to deal with the supervised learning of a multicategory classification problem and serves as a parameter estimator of the neural network classifier. Then, a generalized EM (GEM) algorithm is developed as an alternative to the MLE problem of PMN. The effectiveness of the proposed PMN architecture and developed EM algorithms are assessed by conducting a set of speaker recognition experiments. It is shown that GEM converges faster than EM to the same solution space.

The problem of acoustic modeling for speech recognizers is addressed. We distinguish two types of speech variability, long term (speaker identity, stationary noise, channel distortion) and short term (phoneme class). Currently, most recognizers model the two variabilities without considering their specificities, which may result in flat distributions with limited discriminability. In our system, the long term variability (environment) is modeled by a mixture model, where each mixture is modeled by a mixture stochastic trajectory model (MSTM). We propose the environment dependent mixture stochastic trajectory model (ED-MSTM) to model a set of environments. The parameters of ED-MSTM are estimated using the maximum likelihood (ML) estimation criterion by the expectation-maximisation (EM) algorithm. Our model has been tested on a 1011 word vocabulary, multi-speaker continuous French recognition task with noisy speech. In the experiments, we assume that speakers can be grouped into a pre-determined number of classes and that the class label of a speaker is missing. The use of environmental modeling cut down the error rate produced by the multi-speaker system by about 15%, which is a statistically significant improvement. The idea of environment modeling is applicable to other acoustic modeling techniques such as hidden Markov models.

Source localisation is among the most important steps in array processing. The authors present a neuromimetic approach in signal processing. A set of neural networks is used to find the azimuth of one or several sources impinging on a linear array of equally spaced sensors. Each network in this set is specialised to determine if there is an emitter in a given angular sector. Thus a neural network has a specific architecture suited to detect and enhance the signal coming from the angular sector it is associated with. The performances of this method on real underwater signals confirm the encouraging results obtained on simulation tests. >

The principle of trajectory space comparison for text-independent speaker recognition and some solutions to the space comparison problem based on vector quantization are presented. The comparison of the recognition rates of different solutions is reported. The experimental system achieves a 99.5% text-independent speaker recognition rate for 23 speakers, using five phrases for training and five for test. A speaker-independent continuous speech recognition system is built in which this principle is used for speaker adaptation. >

The spectrogram displays the energy distribution in the signal as a function of both time and frequency. High performance phonetic labeling of spectrogram can be obtained without the use of higher level knowledge sources. This paper reports on an attempt to acquire a phonetic knowledge base about Arabic amphatic consonants in order to build an automatic speech recognition system for modern Arabic language. The acquisition of this knowledge base is done by analysing the spectrogram of these consonants in multiple context.

We describe a continuous speech recognition system that uses the multi-band paradigm. This principle is based on the recombination of several independent sub-recognizers, each one assigned to a specific frequency band. The major issue of such systems consists of deciding at which time the recombination must be done. Our algorithm lets each band be totally independent from the others, and uses the different solutions to resegment the initial sentence. Finally, the bands are synchronously merged together, according to this new segmentation. The whole system is too complex to be entirely described here, and, in this paper, we concentrate on the synchronous recombination part, which is achieved by a classifier. The system has been tested in clean and noisy environments, and proved to be especially robust to noise.

The authors deals with the issue of spectral line tracking in a time/frequency sonar image, the so-called lofargram (LOw Frequency Analysis Representation). The approach, based on feature grouping theory, consists of studying the line notion with respect to the minimization of a cost function Phi . A multistage decision process is used to extract optimal paths from an image window. These optimal paths are defined according to Phi and have a high chance of belonging to a spectral line. Then, a temporal processing integrates these observation windows to supply, after filtering, an image cleared of locally incoherent noise. Thus, the spectral line perception is drastically enhanced, even with a low signal to noise ratio. >

The authors address the problem of speaker recognition using very short utterances, both for training and for recognition. The authors propose to exploit speaker-specific correlations between two suitably defined parameter vector sequences. A nonlinear vectorial interpolation technique is used to capture speaker-specific information, through least-square-error minimization. The experiments show the feasibility of recognizing a speaker among a population of about 100 persons using only an utterance of one word both for training and for recognition. >

The authors present an NN-IIR (neural network/infinite impulse response filter) system for phoneme recognition in continuous speech based on the idea of modeling the recognition process by state evolution and interpretation equations. This work gives a solution to temporal information representation in phoneme recognition using neural networks and recursive filters, yielding better recognition results for continuous speech. This recognition system has two promising properties, i.e., capabilities for dealing with sequential properties and for interpreting speech signals by means of a training process. It was shown experimentally that the NN-IIR network obtained good performance for continuous speech recognition. Preliminary experiments with limited training data indicate that the NN-IIR provides good discrimination power for plosives, which are highly context-dependent. >

Presents the basic principles and some preliminary applications of a novel connectionist model for building up a neural network system. The model uses the cortical column instead of the classical neuron as the basic processing unit; it is closely related to the neurobiological modeling of the human cortex. The main features of this model which make it different from classical approaches concern the local connectivity and the integration of time by causality learning. The model provides a basic unit well -adapted to solve humanlike problems by integrating particular difficulties in its own structure (e.g. the coarticulation effect in speech recognition). The model has been applied to two difficult problems in the artificial intelligence field: speech recognition (more precisely, the acoustic-phonetic decoding of continuous speech) and biomedical X-ray image interpretation. The two systems that were designed for these applications demonstrate the ability of the cortical column to solve perceptive and cognitive tasks