Speaker identification using higher order spectral phase features and their effectiveness vis-a-vis Mel-Cepstral features

The effectiveness of higher-order spectral (HOS) phase features in speaker recognition is investigated by comparison with Mel Cepstral features on the same speech data. HOS phase features retain phase information from the Fourier spectrum unlikeMel–frequency Cepstral coefficients (MFCC). Gaussian mixture models are constructed from Mel– Cepstral features and HOS features, respectively, for the same data from various speakers in the Switchboard telephone Speech Corpus. Feature clusters, model parameters and classification performance are analyzed. HOS phase features on their own provide a correct identification rate of about 97% on the chosen subset of the corpus. This is the same level of accuracy as provided by MFCCs. Cluster plots and model parameters are compared to show that HOS phase features can provide complementary information to better discriminate between speakers.

Spectral coding methods for speech compression and speaker identification

This thesis investigates aspects of encoding the speech spectrum at low bit rates, with extensions to the effect of such coding on automatic speaker identification. Vector quantization (VQ) is a technique for jointly quantizing a block of samples at once, in order to reduce the bit rate of a coding system. The major drawback in using VQ is the complexity of the encoder. Recent research has indicated the potential applicability of the VQ method to speech when product code vector quantization (PCVQ) techniques are utilized. The focus of this research is the efficient representation, calculation and utilization of the speech model as stored in the PCVQ codebook. In this thesis, several VQ approaches are evaluated, and the efficacy of two training algorithms is compared experimentally. It is then shown that these productcode vector quantization algorithms may be augmented with lossless compression algorithms, thus yielding an improved overall compression rate. An approach using a statistical model for the vector codebook indices for subsequent lossless compression is introduced. This coupling of lossy compression and lossless compression enables further compression gain. It is demonstrated that this approach is able to reduce the bit rate requirement from the current 24 bits per 20 millisecond frame to below 20, using a standard spectral distortion metric for comparison. Several fast-search VQ methods for use in speech spectrum coding have been evaluated. The usefulness of fast-search algorithms is highly dependent upon the source characteristics and, although previous research has been undertaken for coding of images using VQ codebooks trained with the source samples directly, the product-code structured codebooks for speech spectrum quantization place new constraints on the search methodology. The second major focus of the research is an investigation of the effect of lowrate spectral compression methods on the task of automatic speaker identification. The motivation for this aspect of the research arose from a need to simultaneously preserve the speech quality and intelligibility and to provide for machine-based automatic speaker recognition using the compressed speech. This is important because there are several emerging applications of speaker identification where compressed speech is involved. Examples include mobile communications where the speech has been highly compressed, or where a database of speech material has been assembled and stored in compressed form. Although these two application areas have the same objective - that of maximizing the identification rate - the starting points are quite different. On the one hand, the speech material used for training the identification algorithm may or may not be available in compressed form. On the other hand, the new test material on which identification is to be based may only be available in compressed form. Using the spectral parameters which have been stored in compressed form, two main classes of speaker identification algorithm are examined. Some studies have been conducted in the past on bandwidth-limited speaker identification, but the use of short-term spectral compression deserves separate investigation. Combining the major aspects of the research, some important design guidelines for the construction of an identification model when based on the use of compressed speech are put forward.

An approach to statistical lip modelling for speaker identification via chromatic feature extraction

This paper presents a novel technique for the tracking of moving lips for the purpose of speaker identification. In our system, a model of the lip contour is formed directly from chromatic information in the lip region. Iterative refinement of contour point estimates is not required. Colour features are extracted from the lips via concatenated profiles taken around the lip contour. Reduction of order in lip features is obtained via principal component analysis (PCA) followed by linear discriminant analysis (LDA). Statistical speaker models are built from the lip features based on the Gaussian mixture model (GMM). Identification experiments performed on the M2VTS¹ database, show encouraging results

The use of temporal speech and lip information for multi-modal speaker identification via multi-stream HMMs

Investigates the use of temporal lip information, in conjunction with speech information, for robust, text-dependent speaker identification. We propose that significant speaker-dependent information can be obtained from moving lips, enabling speaker recognition systems to be highly robust in the presence of noise. The fusion structure for the audio and visual information is based around the use of multi-stream hidden Markov models (MSHMM), with audio and visual features forming two independent data streams. Recent work with multi-modal MSHMMs has been performed successfully for the task of speech recognition. The use of temporal lip information for speaker identification has been performed previously (T.J. Wark et al., 1998), however this has been restricted to output fusion via single-stream HMMs. We present an extension to this previous work, and show that a MSHMM is a valid structure for multi-modal speaker identification

Design of speaker identification schemes for large number of speakers (A)

Design of speaker identification schemes for a small number of speakers (around 10) with a high degree of accuracy in controlled environment is a practical proposition today. When the number of speakers is large (say 50–100), many of these schemes cannot be directly extended, as both recognition error and computation time increase monotonically with population size. The feature selection problem is also complex for such schemes. Though there were earlier attempts to rank order features based on statistical distance measures, it has been observed only recently that the best two independent measurements are not the same as the combination in two's for pattern classification. We propose here a systematic approach to the problem using the decision tree or hierarchical classifier with the following objectives: (1) Design of optimal policy at each node of the tree given the tree structure i.e., the tree skeleton and the features to be used at each node. (2) Determination of the optimal feature measurement and decision policy given only the tree skeleton. Applicability of optimization procedures such as dynamic programming in the design of such trees is studied. The experimental results deal with the design of a 50 speaker identification scheme based on this approach.

Voice source characterization using pitch synchronous discrete cosine transform for speaker identification

A characterization of the voice source (VS) signal by the pitch synchronous (PS) discrete cosine transform (DCT) is proposed. With the integrated linear prediction residual (ILPR) as the VS estimate, the PS DCT of the ILPR is evaluated as a feature vector for speaker identification (SID). On TIMIT and YOHO databases, using a Gaussian mixture model (GMM)-based classifier, it performs on par with existing VS-based features. On the NIST 2003 database, fusion with a GMM-based classifier using MFCC features improves the identification accuracy by 12% in absolute terms, proving that the proposed characterization has good promise as a feature for SID studies. (C) 2015 Acoustical Society of America

Study on Closed-set Speaker Identification Based on Biomimetic Pattern Recognition

In this paper, a new classifier of speaker identification has been proposed, which is based on Biomimetic pattern recognition (BPR). Distinguished from traditional speaker recognition methods, such as DWT, HMM, GMM, SVM and so on, the proposed classifier is constructed by some finite sub-space which is reasonable covering of the points in high dimensional space according to distributing characteristic of speech feature points. It has been used in the system of speaker identification. Experiment results show that better effect could be obtained especially with lesser samples. Furthermore, the proposed classifier employs a much simpler modeling structure as compared to the GMM. In addition, the basic idea "cognition" of Biomimetic pattern recognition (BPR) results in no requirement of retraining the old system for enrolling new speakers.

Speaker identification using models for phonemes

Motivation for Speaker recognition work is presented in the first part of the thesis. An exhaustive survey of past work in this field is also presented. A low cost system not including complex computation has been chosen for implementation. Towards achieving this a PC based system is designed and developed. A front end analog to digital convertor (12 bit) is built and interfaced to a PC. Software to control the ADC and to perform various analytical functions including feature vector evaluation is developed. It is shown that a fixed set of phrases incorporating evenly balanced phonemes is aptly suited for the speaker recognition work at hand. A set of phrases are chosen for recognition. Two new methods are adopted for the feature evaluation. Some new measurements involving a symmetry check method for pitch period detection and ACE‘ are used as featured. Arguments are provided to show the need for a new model for speech production. Starting from heuristic, a knowledge based (KB) speech production model is presented. In this model, a KB provides impulses to a voice producing mechanism and constant correction is applied via a feedback path. It is this correction that differs from speaker to speaker. Methods of defining measurable parameters for use as features are described. Algorithms for speaker recognition are developed and implemented. Two methods are presented. The first is based on the model postulated. Here the entropy on the utterance of a phoneme is evaluated. The transitions of voiced regions are used as speaker dependent features. The second method presented uses features found in other works, but evaluated differently. A knock—out scheme is used to provide the weightage values for the selection of features. Results of implementation are presented which show on an average of 80% recognition. It is also shown that if there are long gaps between sessions, the performance deteriorates and is speaker dependent. Cross recognition percentages are also presented and this in the worst case rises to 30% while the best case is 0%. Suggestions for further work are given in the concluding chapter.

A naïve, salience-based method for speaker identification in fiction books

This paper presents a salience-based technique for the annotation of directly quoted speech from fiction text. In particular, this paper determines to what extent a naïve (without the use of complex machine learning or knowledge-based techniques) scoring technique can be used for the identification of the speaker of speech quotes. The presented technique makes use of a scoring technique, similar to that commonly found in knowledge-poor anaphora resolution research, as well as a set of hand-coded rules for the final identification of the speaker of each quote in the text. Speaker identification is shown to be achieved using three tasks: the identification of a speech-verb associated with a quote with a recall of 94.41%; the identification of the actor associated with a quote with a recall of 88.22%; and the selection of a speaker with an accuracy of 79.40%.

Hierarchical rule generalisation for speaker identification in fiction books

This paper presents a hierarchical pattern matching and generalisation technique which is applied to the problem of locating the correct speaker of quoted speech found in fiction books. Patterns from a training set are generalised to create a small number of rules, which can be used to identify items of interest within the text. The pattern matching technique is applied to finding the Speech-Verb, Actor and Speaker of quotes found in ction books. The technique performs well over the training data, resulting in rule-sets many times smaller than the training set, but providing very high accuracy. While the rule-set generalised from one book is less effective when applied to different books than an approach based on hand coded heuristics, performance is comparable when testing on data closely related to the training set.