Speaker Identification Performance Research Articles

Speaker identification investigation and analysis in unbiased and biased emotional talking environments

This work aims at investigating and analyzing speaker identification in each unbiased and biased emotional talking environments based on a classifier called Suprasegmental Hidden Markov Models (SPHMMs). The first talking environment is unbiased towards any emotion, while the second talking environment is biased towards different emotions. Each of these talking environments is made up of six distinct emotions. These emotions are neutral, angry, sad, happy, disgust and fear. The investigation and analysis of this work show that speaker identification performance in the biased talking environment is superior to that in the unbiased talking environment. The obtained results in this work are close to those achieved in subjective assessment by human judges.

Neural network based feature transformation for emotion independent speaker identification

In this paper we are proposing neural network based feature transformation framework for developing emotion independent speaker identification system. Most of the present speaker recognition systems may not perform well during emotional environments. In real life, humans extensively express emotions during conversations for effectively conveying the messages. Therefore, in this work we propose the speaker recognition system, robust to variations in emotional moods of speakers. Neural network models are explored to transform the speaker specific spectral features from any specific emotion to neutral. In this work, we have considered eight emotions namely, Anger, Sad, Disgust, Fear, Happy, Neutral, Sarcastic and Surprise. The emotional databases developed in Hindi, Telugu and German are used in this work for analyzing the effect of proposed feature transformation on the performance of speaker identification system. In this work, spectral features are represented by mel-frequency cepstral coefficients, and speaker models are developed using Gaussian mixture models. Performance of the speaker identification system is analyzed with various feature mapping techniques. Results have demonstrated that the proposed neural network based feature transformation has improved the speaker identification performance by 20 %. Feature transformation at the syllable level has shown the better performance, compared to sentence level.

Hierarchical Speaker Identification based on Latent Variable Decomposition

paper, a novel hierarchical speaker identification method based on Latent Variable Decomposition (LVD) has been proposed. Firstly, we got a coarse decision by a fast scan all registered speakers using LVD based features and GMM classifier to find R possible target speakers, and then MFCC or PCA based features were used to make final decision. LVD has another advantage: reduction of the feature vectors dimensions, and the noise is removed from speech simultaneity. So, it can reduce the computational complexity and improve the performance of speaker identification. The experimental results showed that the proposed method could improve recognition accuracy of system remarkably and the system has better robustness by comparing with the traditional speaker identification method. General Terms

A technique for adjusting Gaussian mixture model weights that improves speaker identification performance in the presence of phonemic train/test mismatch.

Speaker identification is complicated by cases where training material is phonemically deficient. Misclassifications can result either because subsequent test material from that speaker contains primarily the phonemes missing from the training data or because that test material is phonemically most consistent with another talker’s model. This situation can arise in any dialog where, for reasons of brevity and clarity, conventions must be imposed on phraseology. We present here a technique for detecting phonemic deficiencies in a speaker model, and then correcting that model to partially compensate for the biased training data. This technique relies upon a specially constructed universal background model (UBM) from which speaker models are adapted. This UBM is formed by weighting several dozen phoneme GMMs using EM training. As a result, each Gaussian component of the UBM (and of the resulting speaker models) corresponds to a specific phoneme. Analysis of the speaker model weights reveals whether the training data had the typical phonemic variety found in ordinary speech, and if it did not, the weights are adjusted. Using a specially designed corpus created from the TIMIT utterances, we show that this reweighting technique improves performance over non-reweighted models. Results are also given for the Air Traffic Control Corpus.

Identifying speakers using their emotion cues

This paper addresses the formulation of a new speaker identification approach which employs knowledge of emotional content of speaker information. Our proposed approach in this work is based on a two-stage recognizer that combines and integrates both emotion recognizer and speaker recognizer into one recognizer. The proposed approach employs both Hidden Markov Models (HMMs) and Suprasegmental Hidden Markov Models (SPHMMs) as classifiers. In the experiments, six emotions are considered including neutral, angry, sad, happy, disgust and fear. Our results show that average speaker identification performance based on the proposed two-stage recognizer is 79.92% with a significant improvement over a one-stage recognizer with an identification performance of 71.58%. The results obtained based on the proposed approach are close to those achieved in subjective evaluation by human listeners.

An Overview of Speaker Identification: Accuracy and Robustness Issues

This paper presents the main paradigms for speaker identification, and recent work on missing data methods to increase robustness. The feature extraction, speaker modeling and system classification are discussed. Evaluations of speaker identification performance subject to environmental noise are presented. While performance is impressive in clean speech conditions, there is rapid degradation with mismatched additive noise. Missing data methods can compensate against arbitrary disturbances and remove environmental mismatches. An overview of missing data methods is provided and applications to robust speaker identification summarized. Finally combined approaches involving bottom-up estimation and top-down processing are reviewed, and their significance discussed.

Robust speaker identification in the presence of car noise

A robust feature set, Teager Energy Operator based Cepstral Coefficients (TEOCC) for speaker identification task is proposed in this paper. Admissible Wavelet Packet (AWP) transform and the Teager Energy Operator (TEO) is used to obtain robust features. The proposed features significantly improve the speaker identification performance (76.25%) compared with the Mel Frequency Cepstral Coefficient (MFCC) features (57%) in the presence of car noise. The performance is evaluated using TIMIT and NOISEX-92 databases. This paper shows that higher-frequency bands also carry more speaker-specific information and the identification rate can be improved without additional processing of the signal to remove noise.

Limited data speaker identification

In this paper, the task of identifying the speaker using limited training and testing data is addressed. Speaker identification system is viewed as four stages namely, analysis, feature extraction, modelling and testing. The speaker identification performance depends on the techniques employed in these stages. As demonstrated by different experiments, in case of limited training and testing data condition, owing to less data, existing techniques in each stage will not provide good performance. This work demonstrates the following: multiple frame size and rate (MFSR) analysis provides improvement in the analysis stage, combination of mel frequency cepstral coefficients (MFCC), its temporal derivatives (Δ, ΔΔ), linear prediction residual (LPR) and linear prediction residual phase (LPRP) features provides improvement in the feature extraction stage and combination of learning vector quantization (LVQ) and gaussian mixture model — universal background model (GMM-UBM) provides improvement in the modelling stage. The performance is further improved by integrating the proposed techniques at the respective stages and combining the evidences from them at the testing stage. To achieve this, we propose strength voting (SV), weighted borda count (WBC) and supporting systems (SS) as combining methods at the abstract, rank and measurement levels, respectively. Finally, the proposed hierarchical combination (HC) method integrating these three methods provides significant improvement in the performance. Based on these explorations, thiswork proposes a scheme for speaker identification under limited training and testing data.

Likelihood Ratio Based Score Fusion for Audio-Visual Speaker Identification in Challenging Environment

It is well known to enhance the performance of noise robust speaker identification using visual speech information with audio utterances. This paper presents an approach to evaluate the performance of a noise robust audio-visual speaker identification system using likelihood ratio based score fusion in challenging environment. Though the traditional HMM based audio-visual speaker identification system is very sensitive to the speech parameter variation, the proposed likelihood ratio based score fusion method is found to be stance and performs well for improving the robustness and naturalness of human-computerinteraction. In this paper, we investigate the proposed audiovisual speaker identification system in typical office environments conditions. To do this, we investigated two approaches that utilize speech utterance with visual features to improve speaker identification performance in acoustically and visually challenging environment: one seeks to eliminate the noise from the acoustic and visual features by using speech and facial image pre-processing techniques. The other task combines speech and facial features that have been used by the multiple Discrete Hidden Markov Model classifiers with likelihood ratio based score fusion. It is shown that the proposed system can improve a significant amount of performance for audio-visual speaker identification in challenging official environment conditions.

Employing Second-Order Circular Suprasegmental Hidden Markov Models to Enhance Speaker Identification Performance in Shouted Talking Environments

Speaker identification performance is almost perfect in neutral talking environments; however, the performance is deteriorated significantly in shouted talking environments. This work is devoted to proposing, implementing and evaluating new models called Second-Order Circular Suprasegmental Hidden Markov Models (CSPHMM2s) to alleviate the deteriorated performance in the shouted talking environments. These proposed models possess the characteristics of both Circular Suprasegmental Hidden Markov Models (CSPHMMs) and Second-Order Suprasegmental Hidden Markov Models (SPHMM2s). The results of this work show that CSPHMM2s outperform each of: First-Order Left-to-Right Suprasegmental Hidden Markov Models (LTRSPHMM1s), Second-Order Left-to-Right Suprasegmental Hidden Markov Models (LTRSPHMM2s) and First-Order Circular Suprasegmental Hidden Markov Models (CSPHMM1s) in the shouted talking environments. In such talking environments and using our collected speech database, average speaker identification performance based on LTRSPHMM1s, LTRSPHMM2s, CSPHMM1s and CSPHMM2s is 74.6%, 78.4%, 78.7% and 83.4%, respectively. Speaker identification performance obtained based on CSPHMM2s is close to that obtained based on subjective assessment by human listeners.

Speaker Recognition by Combining MFCC and Phase Information in Noisy Conditions

In this paper, we investigate the effectiveness of phase for speaker recognition in noisy conditions and combine the phase information with mel-frequency cepstral coefficients (MFCCs). To date, almost speaker recognition methods are based on MFCCs even in noisy conditions. For MFCCs which dominantly capture vocal tract information, only the magnitude of the Fourier Transform of time-domain speech frames is used and phase information has been ignored. High complement of the phase information and MFCCs is expected because the phase information includes rich voice source information. Furthermore, some researches have reported that phase based feature was robust to noise. In our previous study, a phase information extraction method that normalizes the change variation in the phase depending on the clipping position of the input speech was proposed, and the performance of the combination of the phase information and MFCCs was remarkably better than that of MFCCs. In this paper, we evaluate the robustness of the proposed phase information for speaker identification in noisy conditions. Spectral subtraction, a method skipping frames with low energy/Signal-to-Noise (SN) and noisy speech training models are used to analyze the effect of the phase information and MFCCs in noisy conditions. The NTT database and the JNAS (Japanese Newspaper Article Sentences) database added with stationary/non-stationary noise were used to evaluate our proposed method. MFCCs outperformed the phase information for clean speech. On the other hand, the degradation of the phase information was significantly smaller than that of MFCCs for noisy speech. The individual result of the phase information was even better than that of MFCCs in many cases by clean speech training models. By deleting unreliable frames (frames having low energy/SN), the speaker identification performance was improved significantly. By integrating the phase information with MFCCs, the speaker identification error reduction rate was about 30%-60% compared with the standard MFCC-based method.

Voice biometric feature using Gammatone filterbank and ICA

Voice biometric feature extraction is the core task in developing any speaker identification system. This paper proposes a robust feature extraction technique for the purpose of speaker identification. The technique is based on processing monaural speech signal using human auditory system based Gammatone Filterbank (GTF) and Independent Component Analysis (ICA). The measures used to assess the robustness to additive noises and speaker identification performance are defined and discussed. The kkn the proposed feature is evaluated in real environments under varying noisy conditions. The proposed feature is benchmarked against the commonly used features such as: MFCC, PLCC, and PLP, and it outperforms them in different noisy environments.

Modulation Spectral Features for Robust Far-Field Speaker Identification

In this paper, auditory inspired modulation spectral features are used to improve automatic speaker identification (ASI) performance in the presence of room reverberation. The modulation spectral signal representation is obtained by first filtering the speech signal with a 23-channel gammatone filterbank. An eight-channel modulation filterbank is then applied to the temporal envelope of each gammatone filter output. Features are extracted from modulation frequency bands ranging from 3-15 H z and are shown to be robust to mismatch between training and testing conditions and to increasing reverberation levels. To demonstrate the gains obtained with the proposed features, experiments are performed with clean speech, artificially generated reverberant speech, and reverberant speech recorded in a meeting room. Simulation results show that a Gaussian mixture model based ASI system, trained on the proposed features, consistently outperforms a baseline system trained on mel-frequency cepstral coefficients. For multimicrophone ASI applications, three multichannel score combination and adaptive channel selection techniques are investigated and shown to further improve ASI performance.

Employing Second-Order Circular Suprasegmental Hidden Markov Models to Enhance Speaker Identification Performance in Shouted Talking Environments

Employing Second-Order Circular Suprasegmental Hidden Markov Models to Enhance Speaker Identification Performance in Shouted Talking Environments

Speaker identification in the shouted environment using Suprasegmental Hidden Markov Models

In this paper, Suprasegmental Hidden Markov Models (SPHMMs) have been used to enhance the recognition performance of text-dependent speaker identification in the shouted environment. Our speech database consists of two databases: our collected database and the Speech Under Simulated and Actual Stress (SUSAS) database. Our results show that SPHMMs significantly enhance speaker identification performance compared to Second-Order Circular Hidden Markov Models (CHMM2s) in the shouted environment. Using our collected database, speaker identification performance in this environment is 68% and 75% based on CHMM2s and SPHMMs, respectively. Using the SUSAS database, speaker identification performance in the same environment is 71% and 79% based on CHMM2s and SPHMMs, respectively.

Mitigation of Nonlinear Distortion in Speech Signals Using Histogram Matching

Nonlinear distortion is a common artifact in audio communication equipment. In addition, it is well know that the normalized amplitude distribution of speech signals converges to approximately a gamma distribution after a few seconds. Hence, the transfer function of any memoryless nonlinearity distorting the speech signal can easily be estimated, provided one has buffered enough data. This research shows how both parametric and nonparametric histogram matching algorithms can be employed to remove the effects of these types of distortions. Further, the improvement these algorithms have on speaker identification performance is also studied. This approach represents a radical departure from the traditional approach taken in speech enhancement. The traditional approach has been to first acquire a model or representation of the distortion, noise, or interference that is corrupting the signal, and then attempt to remove this from the signal, usually introducing other forms of distortion in the process. With this approach a model of clean undistorted speech is used that the distorted speech is then matched to.

External factors impacting the performance of speaker identification: Multisession audio research project (MARP) corpus experiments

This study looks at the effect of data conditions on automated speech processing systems. The goal is to better understand the impact of acoustical features on accuracy and to develop more robust features. A speaker identification (SID) system was used for the experiments. To explore this issue, a new longitudinal database was collected involving 60 speakers over 18 months. This corpus allowed us to examine four data factors that impact SID: (1) intersession variability, (2) question intonation (3), text-dependency (identical phonetic content), and (4) whispered speech. First, we found that intersession SID suffered an average loss in accuracy of 17% independent of the time latency between sessions. Second, differing the intonation conditions in train and test hurt SID performance by 5%. Third, text-dependent data showed the most dramatic impact, where using phonetically-identical test and train sentences yielded 0% error. However, replacing the target speakers content with a random text-independent sentence, ceteris paribus, caused accuracy to plummet 94%. In all of the erroneous identifications, the top-ranked speaker was speaking the sentence used for the training model. Finally, when there was a mismatch between whispered speech being used in training and normally phonated speech in testing, the SID performance was severely impacted.

Enhancing speaker identification performance under the shouted talking condition using second-order circular hidden Markov models

It is known that the performance of speaker identification systems is high under the neutral talking condition; however, the performance deteriorates under the shouted talking condition. In this paper, second-order circular hidden Markov models (CHMM2s) have been proposed and implemented to enhance the performance of isolated-word text-dependent speaker identification systems under the shouted talking condition. Our results show that CHMM2s significantly improve speaker identification performance under such a condition compared to the first-order left-to-right hidden Markov models (LTRHMM1s), second-order left-to-right hidden Markov models (LTRHMM2s), and the first-order circular hidden Markov models (CHMM1s). Under the shouted talking condition, our results show that the average speaker identification performance is 23% based on LTRHMM1s, 59% based on LTRHMM2s, and 60% based on CHMM1s. On the other hand, the average speaker identification performance under the same talking condition based on CHMM2s is 72%.

Model-based sequential organization in cochannel speech

A human listener has the ability to follow a speaker's voice while others are speaking simultaneously; in particular, the listener can organize the time-frequency energy of the same speaker across time into a single stream. In this paper, we focus on sequential organization in cochannel speech, or mixtures of two voices. We extract minimally corrupted segments, or usable speech, in cochannel speech using a robust multipitch tracking algorithm. The extracted usable speech is shown to capture speaker characteristics and improves speaker identification (SID) performance across various target-to-interferer ratios. To utilize speaker characteristics for sequential organization, we extend the traditional SID framework to cochannel speech and derive a joint objective for sequential grouping and SID, leading to a problem of search for the optimum hypothesis. Subsequently we propose a hypothesis pruning algorithm based on speaker models in order to make the search computationally efficient. Evaluation results show that the proposed system approaches the ceiling SID performance obtained with prior pitch information and yields significant improvement over alternative approaches to sequential organization.

Improving Speaker Identification Performance Under the Shouted Talking Condition Using the Second-Order Hidden Markov Models

Speaker identification systems perform well under the neutral talking condition; however, they suffer sharp degradation under the shouted talking condition. In this paper, the second-order hidden Markov models (HMM2s) have been used to improve the recognition performance of isolated-word text-dependent speaker identification systems under the shouted talking condition. Our results show that HMM2s significantly improve the speaker identification performance compared to the first-order hidden Markov models (HMM1s). The average speaker identification performance under the shouted talking condition based on HMM1s is. On the other hand, the average speaker identification performance based on HMM2s is.