Automatic spoken language identification utilizing acoustic and phonetic speech information

Automatic spoken Language Identi¯cation (LID) is the process of identifying the language spoken within an utterance. The challenge that this task presents is that no prior information is available indicating the content of the utterance or the identity of the speaker. The trend of globalization and the pervasive popularity of the Internet will amplify the need for the capabilities spoken language identi¯ca- tion systems provide. A prominent application arises in call centers dealing with speakers speaking di®erent languages. Another important application is to index or search huge speech data archives and corpora that contain multiple languages. The aim of this research is to develop techniques targeted at producing a fast and more accurate automatic spoken LID system compared to the previous National Institute of Standards and Technology (NIST) Language Recognition Evaluation. Acoustic and phonetic speech information are targeted as the most suitable fea- tures for representing the characteristics of a language. To model the acoustic speech features a Gaussian Mixture Model based approach is employed. Pho- netic speech information is extracted using existing speech recognition technol- ogy. Various techniques to improve LID accuracy are also studied. One approach examined is the employment of Vocal Tract Length Normalization to reduce the speech variation caused by di®erent speakers. A linear data fusion technique is adopted to combine the various aspects of information extracted from speech. As a result of this research, a LID system was implemented and presented for evaluation in the 2003 Language Recognition Evaluation conducted by the NIST.

Low-cost hardware speech enhancement for improved speech recognition in automotive environments

Voice recognition is one of the key enablers to reduce driver distraction as in-vehicle systems become more and more complex. With the integration of voice recognition in vehicles, safety and usability are improved as the driver’s eyes and hands are not required to operate system controls. Whilst speaker independent voice recognition is well developed, performance in high noise environments (e.g. vehicles) is still limited. La Trobe University and Queensland University of Technology have developed a low-cost hardware-based speech enhancement system for automotive environments based on spectral subtraction and delay–sum beamforming techniques. The enhancement algorithms have been optimised using authentic Australian English collected under typical driving conditions. Performance tests conducted using speech data collected under variety of vehicle noise conditions demonstrate a word recognition rate improvement in the order of 10% or more under the noisiest conditions. Currently developed to a proof of concept stage there is potential for even greater performance improvement.

Improving the PLDA based speaker verification in limited microphone data conditions

A significant amount of speech data is required to develop a robust speaker verification system, but it is difficult to find enough development speech to match all expected conditions. In this paper we introduce a new approach to Gaussian probabilistic linear discriminant analysis (GPLDA) to estimate reliable model parameters as a linearly weighted model taking more input from the large volume of available telephone data and smaller proportional input from limited microphone data. In comparison to a traditional pooled training approach, where the GPLDA model is trained over both telephone and microphone speech, this linear-weighted GPLDA approach is shown to provide better EER and DCF performance in microphone and mixed conditions in both the NIST 2008 and NIST 2010 evaluation corpora. Based upon these results, we believe that linear-weighted GPLDA will provide a better approach than pooled GPLDA, allowing for the further improvement of GPLDA speaker verification in conditions with limited development data.

Rapid yet accurate speech indexing using dynamic match lattice spotting

The support for typically out-of-vocabulary query terms such as names, acronyms, and foreign words is an important requirement of many speech indexing applications. However, to date many unrestricted vocabulary indexing systems have struggled to provide a balance between good detection rate and fast query speeds. This paper presents a fast and accurate unrestricted vocabulary speech indexing technique named Dynamic Match Lattice Spotting (DMLS). The proposed method augments the conventional lattice spotting technique with dynamic sequence matching, together with a number of other novel algorithmic enhancements, to obtain a system that is capable of searching hours of speech in seconds while maintaining excellent detection performance

A finite-rate-of-innovation signal sampling perspective of the source-filter model of speech production

We consider the speech production mechanism and the asso- ciated linear source-filter model. For voiced speech sounds in particular, the source/glottal excitation is modeled as a stream of impulses and the filter as a cascade of second-order resonators. We show that the process of sampling speech signals can be modeled as filtering a stream of Dirac impulses (a model for the excitation) with a kernel function (the vocal tract response),and then sampling uniformly. We show that the problem of esti- mating the excitation is equivalent to the problem of recovering a stream of Dirac impulses from samples of a filtered version. We present associated algorithms based on the annihilating filter and also make a comparison with the classical linear prediction technique, which is well known in speech analysis. Results on synthesized as well as natural speech data are presented.

RIESZ-TRANSFORM-BASED DEMODULATION OF NARROWBAND SPECTROGRAMS OF VOICED SPEECH

Narrowband spectrograms of voiced speech can be modeled as an outcome of two-dimensional (2-D) modulation process. In this paper, we develop a demodulation algorithm to estimate the 2-D amplitude modulation (AM) and carrier of a given spectrogram patch. The demodulation algorithm is based on the Riesz transform, which is a unitary, shift-invariant operator and is obtained as a 2-D extension of the well known 1-D Hilbert transform operator. Existing methods for spectrogram demodulation rely on extension of sinusoidal demodulation method from the communications literature and require precise estimate of the 2-D carrier. On the other hand, the proposed method based on Riesz transform does not require a carrier estimate. The proposed method and the sinusoidal demodulation scheme are tested on real speech data. Experimental results show that the demodulated AM and carrier from Riesz demodulation represent the spectrogram patch more accurately compared with those obtained using the sinusoidal demodulation. The signal-to-reconstruction error ratio was found to be about 2 to 6 dB higher in case of the proposed demodulation approach.

Demodulation of Narrowband Speech Spectrograms Using the Riesz Transform

We propose a two-dimensional (2-D) multicomponent amplitude-modulation, frequency-modulation (AM-FM) model for a spectrogram patch corresponding to voiced speech, and develop a new demodulation algorithm to effectively separate the AM, which is related to the vocal tract response, and the carrier, which is related to the excitation. The demodulation algorithm is based on the Riesz transform and is developed along the lines of Hilbert-transform-based demodulation for 1-D AM-FM signals. We compare the performance of the Riesz transform technique with that of the sinusoidal demodulation technique on real speech data. Experimental results show that the Riesz-transform-based demodulation technique represents spectrogram patches accurately. The spectrograms reconstructed from the demodulated AM and carrier are inverted and the corresponding speech signal is synthesized. The signal-to-noise ratio (SNR) of the reconstructed speech signal, with respect to clean speech, was found to be 2 to 4 dB higher in case of the Riesz transform technique than the sinusoidal demodulation technique.

Automatic transcription of conversational telephone speech

This paper discusses the Cambridge University HTK (CU-HTK) system for the automatic transcription of conversational telephone speech. A detailed discussion of the most important techniques in front-end processing, acoustic modeling and model training, language and pronunciation modeling are presented. These include the use of conversation side based cepstral normalization, vocal tract length normalization, heteroscedastic linear discriminant analysis for feature projection, minimum phone error training and speaker adaptive training, lattice-based model adaptation, confusion network based decoding and confidence score estimation, pronunciation selection, language model interpolation, and class based language models. The transcription system developed for participation in the 2002 NIST Rich Transcription evaluations of English conversational telephone speech data is presented in detail. In this evaluation the CU-HTK system gave an overall word error rate of 23.9%, which was the best performance by a statistically significant margin. Further details on the derivation of faster systems with moderate performance degradation are discussed in the context of the 2002 CU-HTK 10 × RT conversational speech transcription system. © 2005 IEEE.

BOLTZMANN MACHINES FOR SPEECH RECOGNITION.

Boltzmann machines offer a new and exciting approach to automatic speech recognition, and provide a rigorous mathematical formalism for parallel computing arrays. In this paper we briefly summarize Boltzmann machine theory, and present results showing their ability to recognize both static and time-varying speech patterns. A machine with 2000 units was able to distinguish between the 11 steady-state vowels in English with an accuracy of 85%. The stability of the learning algorithm and methods of preprocessing and coding speech data before feeding it to the machine are also discussed. A new type of unit called a carry input unit, which involves a type of state-feedback, was developed for the processing of time-varying patterns and this was tested on a few short sentences. Use is made of the implications of recent work into associative memory, and the modelling of neural arrays to suggest a good configuration of Boltzmann machines for this sort of pattern recognition.

The modified Kanerva model for automatic speech recognition

A parallel processing network derived from Kanerva's associative memory theory Kanerva 1984 is shown to be able to train rapidly on connected speech data and recognize further speech data with a label error rate of 0·68%. This modified Kanerva model can be trained substantially faster than other networks with comparable pattern discrimination properties. Kanerva presented his theory of a self-propagating search in 1984, and showed theoretically that large-scale versions of his model would have powerful pattern matching properties. This paper describes how the design for the modified Kanerva model is derived from Kanerva's original theory. Several designs are tested to discover which form may be implemented fastest while still maintaining versatile recognition performance. A method is developed to deal with the time varying nature of the speech signal by recognizing static patterns together with a fixed quantity of contextual information. In order to recognize speech features in different contexts it is necessary for a network to be able to model disjoint pattern classes. This type of modelling cannot be performed by a single layer of links. Network research was once held back by the inability of single-layer networks to solve this sort of problem, and the lack of a training algorithm for multi-layer networks. Rumelhart, Hinton & Williams 1985 provided one solution by demonstrating the "back propagation" training algorithm for multi-layer networks. A second alternative is used in the modified Kanerva model. A non-linear fixed transformation maps the pattern space into a space of higher dimensionality in which the speech features are linearly separable. A single-layer network may then be used to perform the recognition. The advantage of this solution over the other using multi-layer networks lies in the greater power and speed of the single-layer network training algorithm. © 1989.

Speaker Recognition In Noisy Conditions With Limited Training Data

In this paper we present a novel method for performing speaker recognition with very limited training data and in the presence of background noise. Similarity-based speaker recognition is considered so that speaker models can be created with limited training speech data. The proposed similarity is a form of cosine similarity used as a distance measure between speech feature vectors. Each speech frame is modelled using subband features, and into this framework, multicondition training and optimal feature selection are introduced, making the system capable of performing speaker recognition in the presence of realistic, time-varying noise, which is unknown during training. Speaker identi?cation experiments were carried out using the SPIDRE database. The performance of the proposed new system for noise compensation is compared to that of an oracle model; the speaker identi?cation accuracy for clean speech by the new system trained with limited training data is compared to that of a GMM trained with several minutes of speech. Both comparisons have demonstrated the effectiveness of the new model. Finally, experiments were carried out to test the new model for speaker identi?cation given limited training data and with differing levels and types of realistic background noise. The results have demonstrated the robustness of the new system.

Speech Analysis using Modern Techniques of Nonlinear Dynamics

Medical fields requires fast, simple and noninvasive methods of diagnostic techniques. Several methods are available and possible because of the growth of technology that provides the necessary means of collecting and processing signals. The present thesis details the work done in the field of voice signals. New methods of analysis have been developed to understand the complexity of voice signals, such as nonlinear dynamics aiming at the exploration of voice signals dynamic nature. The purpose of this thesis is to characterize complexities of pathological voice from healthy signals and to differentiate stuttering signals from healthy signals. Efficiency of various acoustic as well as non linear time series methods are analysed. Three groups of samples are used, one from healthy individuals, subjects with vocal pathologies and stuttering subjects. Individual vowels/ and a continuous speech data for the utterance of the sentence "iruvarum changatimaranu" the meaning in English is "Both are good friends" from Malayalam language are recorded using a microphone . The recorded audio are converted to digital signals and are subjected to analysis.Acoustic perturbation methods like fundamental frequency (FO), jitter, shimmer, Zero Crossing Rate(ZCR) were carried out and non linear measures like maximum lyapunov exponent(Lamda max), correlation dimension (D2), Kolmogorov exponent(K2), and a new measure of entropy viz., Permutation entropy (PE) are evaluated for all three groups of the subjects. Permutation Entropy is a nonlinear complexity measure which can efficiently distinguish regular and complex nature of any signal and extract information about the change in dynamics of the process by indicating sudden change in its value. The results shows that nonlinear dynamical methods seem to be a suitable technique for voice signal analysis, due to the chaotic component of the human voice. Permutation entropy is well suited due to its sensitivity to uncertainties, since the pathologies are characterized by an increase in the signal complexity and unpredictability. Pathological groups have higher entropy values compared to the normal group. The stuttering signals have lower entropy values compared to the normal signals.PE is effective in charaterising the level of improvement after two weeks of speech therapy in the case of stuttering subjects. PE is also effective in characterizing the dynamical difference between healthy and pathological subjects. This suggests that PE can improve and complement the recent voice analysis methods available for clinicians. The work establishes the application of the simple, inexpensive and fast algorithm of PE for diagnosis in vocal disorders and stuttering subjects.

Speech sample estimation from composite zerocrossings and encoding via adaptive switching of transforms

This thesis investigates the potential use of zerocrossing information for speech sample estimation. It provides 21 new method tn) estimate speech samples using composite zerocrossings. A simple linear interpolation technique is developed for this purpose. By using this method the A/D converter can be avoided in a speech coder. The newly proposed zerocrossing sampling theory is supported with results of computer simulations using real speech data. The thesis also presents two methods for voiced/ unvoiced classification. One of these methods is based on a distance measure which is a function of short time zerocrossing rate and short time energy of the signal. The other one is based on the attractor dimension and entropy of the signal. Among these two methods the first one is simple and reguires only very few computations compared to the other. This method is used imtea later chapter to design an enhanced Adaptive Transform Coder. The later part of the thesis addresses a few problems in Adaptive Transform Coding and presents an improved ATC. Transform coefficient with maximum amplitude is considered as ‘side information’. This. enables more accurate tfiiz assignment enui step—size computation. A new bit reassignment scheme is also introduced in this work. Finally, sum ATC which applies switching between luiscrete Cosine Transform and Discrete Walsh-Hadamard Transform for voiced and unvoiced speech segments respectively is presented. Simulation results are provided to show the improved performance of the coder