Automatic Speaker Recognition by Speech Signal

For communication, speech is one of the natural forms. A person’s voice contains various parameters that convey information such as emotions, gender, attitude, health, and identity. Determination of these parameters will help to further develop the technology into a reliable and consistent means of identification using speaker recognition. Speaker recognition technologies have wide application areas especially in authentication; surveillance and forensic speaker recognition. In addition, speaker recognition refers to the automated method of identifying or confirming the identity of an individual based on his/her voice. Speech recognition strips out the personal differences to detect the words. Speaker recognition typically disregards the language and meaning to detect the physical person behind the speech. Speech recognition is language-dependent, while Speaker recognition is independent of language. In essence, voice biometrics provides speaker recognition rather than speech recognition. The most accepted form of identification for a human is his/her speech signal. Principally the speaker recognition is the computing task of validating a user’s claimed identity 108using characteristics extracted from their voice. The speaker recognition process based on a speech signal is treated as one of the most exciting technologies of human recognition. For Speaker identification activities we mainly emphasize the physical features of signal. Speakers could be categorized as speaker identification and speaker verification. In speaker identification, the obtained features are compared with all the speaker’s features which are stored in a voice model database and in speaker verification the obtained features are only compared with the stored features of the speaker he/she claimed to be. In this chapter the general principles of speaker recognition, methodology, and applications are discussed.

Human-computer interaction for virtual-real fusion

This paper describes the design and implementation of a realtime speaker recognition system. The system performs text independent, closed set speaker recognition with up to 30 talkers in realtime. In addition, the reference speech used to characterize the 30 talkers can be extracted from as little as 10 seconds of speech from each talker, and the actual recognition performed with less than one minute of speech from the unknown talker. Two speaker recognition algorithms previously developed by Markel and Pfeifer were investigated for use in the realtime system. The results of this investigation clearly show that Markel's technique is superior for applications using very short speech segments for both the speaker models and the recognition trials. Markel's technique was implemented in realtime in a high speed progranmable signal processor. A test of this implementation with a set of 30 male speakers resulted in recognition accuracies of 93-100% for models generated with only 10 seconds of speech, and recognition trials using only 10 seconds of unknown speech.

Biometric template protection of speech signals and information hiding in speech signals are two challenging issues. To resolve such limitations and increase the level of security, our objective is to build multi-level security systems based on speech signals. So, speech watermarking is used simultaneously with automatic speaker identification. The speech watermarking is performed to embed images into the speech signals that are used for speaker identification. The watermark is extracted for authentication, and then the effect of watermark removal on the performance of the speaker identification system in the presence of degradations is studied. This paper presents an approach for speech watermarking based on empirical mode decomposition (EMD) in different transform domains and singular value decomposition (SVD). The speech signal is decomposed in different transform domains with EMD to yield zero-mean components called intrinsic mode functions (IMFs). The watermark is inserted into one of these IMF components with SVD. A comparison between different transform domains for implementing the proposed watermarking scheme on different IMFs is presented. The log-likelihood ratio (LLR), correlation coefficient (Cr), signal-to-noise ratio (SNR), and spectral distortion (SD) are used as metrics for the comparison. According to the simulation results, we find that the watermark embedding in the discrete sine transform domain provides higher SNR and Cr values and lower SD and LLR values. The proposed approach is robust to different attacks.

Currently, the use of speech recognition is increaseingly in many applications such as mobile device interaction, interactive voice response system, voice search, voice dictation and voice identification. The heart of such applications is speech features needed to represent input signals. However, in real applications, speech signals are sampled with various sampling rates. The different sampling rates of input speech lead to the different features. This makes the speech recognition rate dropping. Therefore, this paper proposes an independent resolution descriptor based on fractal codes obtained by fractal encoding and decoding processes. The encoding process extracts fractal codes from partitioned speech signals, whereas the decoding process reconstructs independent resolution speech signals from the fractal codes. This method can effectively reconstruct speech signals at any sampling rates, especially at a higher sampling rate, which is a grand challenge. The proposed method is evaluated the performance by testing with AN4 corpus of CMU Sphinx speech recognition engine. The experimental results show that the proposed method can improve the accuracy of speech recognition, even if the sampling rate of testing speeches differs from that of training speeches.

A model-based framework of classification error rate estimation is proposed for speech and speaker recognition. It aims at predicting the run-time performance of a hidden Markov model (HMM) based recognition system for a given task vocabulary and grammar without the need of running recognition experiments using a separate set of testing samples. This is highly desirable both in theory and in practice. However, the error rate expression in HMM-based speech recognition systems has no closed form solution due to the complexity of the multi-class comparison process and the need for dynamic time warping to handle various speech patterns. To alleviate the difficulty, we propose a one-dimensional model-based misclassification measure to evaluate the distance between a particular model of interest and a combination of many of its competing models. The error rate for a class characterized by the HMM is then the value of a smoothed zero-one error function given the misclassification measure. The overall error rate of the task vocabulary could then be computed as a function of all the available class error rates. The key here is to evaluate the misclassification measure in terms of the parameters of environmental-matched models without running recognition experiments, where the models are adapted by very limited data that could be just the testing utterance itself. In this paper, we show how the misclassification measure could be approximated by first computing the distance between two mixture Gaussian densities, then between two HMMs with mixture Gaussian state observation densities and finally between two sequences of HMMs. The misclassification measure is then converted into classification error rate. When comparing the error rate obtained in running actual experiments and that of the new framework, the proposed algorithm accurately estimates the classification error rate for many types of speech and speaker recognition problems. Based on the same framework, it is also demonstrated that the error rate of a recognition system in a noisy environment could also be predicted.

The automatic identification of person's identity from their voice is a part of modern telecommunication services. In order to execute the identification task, speech signal has to be transmitted to a remote server. So a performance of the recognition/identification system can be influenced by various distortions that occur when transmitting speech signal through a communication channel. This paper studies an effect of telecommunication channel, particularly commonly used narrowband (NB) speech codecs in current telecommunication networks, on a performance of automatic speaker recognition in the context of a channel/codec mismatch between enrollment and test utterances. An influence of speech coding on speaker identification is assessed by using the reference GMM-UBM method. The results show that the partially mismatched scenario offers better results than the fully matched scenario when speaker recognition is done on speech utterances degraded by the different NB codecs. Moreover, deploying EVS and G.711 codecs in a training process of the recognition system provides the best success rate in the fully mismatched scenario. It should be noted here that the both EVS and G.711codecs offer the best speech quality among the codecs deployed in this study. This finding also fully corresponds with the finding presented by Janicki & Staroszczyk in [1] focusing on other speech codecs.

Speech and speaker recognition technology has made very significant progress in the past 50 years. The progress can be summarized by the following changes: (1) from template matching to corpus-base statistical modeling, e.g., HMM and n-grams, (2) from filter bank/spectral resonance to Cepstral features (Cepstrum + DCepstrum + DDCepstrum), (3) from heuristic time-normalization to DTW/DP matching, (4) from gdistanceh-based to likelihood-based methods, (5) from maximum likelihood to discriminative approach, e.g., MCE/GPD and MMI, (6) from isolated word to continuous speech recognition, (7) from small vocabulary to large vocabulary recognition, (8) from context-independent units to context-dependent units for recognition, (9) from clean speech to noisy/telephone speech recognition, (10) from single speaker to speaker-independent/adaptive recognition, (11) from monologue to dialogue/conversation recognition, (12) from read speech to spontaneous speech recognition, (13) from recognition to understanding, (14) from single-modality (audio signal only) to multi-modal (audio/visual) speech recognition, (15) from hardware recognizer to software recognizer, and (16) from no commercial application to many practical commercial applications. Most of these advances have taken place in both the fields of speech recognition and speaker recognition. The majority of technological changes have been directed toward the purpose of increasing robustness of recognition, including many other additional important techniques not noted above.

Speech feature extraction is the most significant step in any Automatic speaker recognition system. In the last 60 years a lot of research has gone into parametric representation of these speech features. Several techniques are currently being used for Automatic Speaker Recognition. Yet Automatic Speaker Recognition still remains a confront mainly due to variations in speaker's vocal tract with time and health, varying environmental conditions, disparities in the behavior and quality of speech recorders etc. MFCC is a extensively used technique in Automatic speaker recognition. In this paper the performance of MFCC technique was evaluated in a quiet environment. A speaker database containing 30 male and 30 female speakers was created. Two separate experiments were conducted for the performance evaluation of MFCC technique when applied to K means clustering. In the first case the speech features were directly matched. In the second case a VQ codebook was created by clustering the training features of these 60 speakers. A distortion easure based on the minimum Euclidean distance was used for speaker recognition. The failure rate of speaker recognition in first ase was found to be was found to be 10% while in the second case as found to be 14%. Matlab-7.10.0 was used for this study.

The main goal of automatic speech and speaker recognition (ASSR) is to transcribe natural speech and recognize its speaker. Recognizing a spoken sentence is obviously a knowledgeintensive process, which must take into account all variable information about the speech communication process. Out of several approaches, artificial intelligence approach has been observed to give remarkable results by visualizing, analysing and finally making a decision on the acoustic patterns. The present chapter deals with two basic processes for ASSR: modelling process and understanding process. The modelling process involves three stages: enhancement, segmentation and pre-processing. The understanding phase involves the recognition of the speech and the speaker through the known knowledge-based model. It has been observed from the literature that, in any speech processing system, because of channel coupling, many source signals mix together. Thus there may be a chance of variability in speech signals affecting the performance of the speech systems due to some factors like: background and channel noise, electrical noise from different sources, meaningless sounds (a sneeze) or filler words (‘uh’ or ‘um’) between words, different speaking rate, mood and styles of speakers. Due to this some signals may not be observed and may result to out-of-vocabulary (OOV) word problem during recognition or understanding process. Thus speech enhancement has played a vital role in the development of a perfect acoustic model or noise – free artificial word model (AWM) and vowel diphthong model (VDM). To process for speech enhancement in frequency – domain, the speech signal has to be split into frames using blind signal separation (BSS) method (Yilmaz and Richard 2004; Yamashita and Hirai 2004; Araki et al 2002; Murata et al 2001; G. J. Jang and Lee 2003; F. Bach and Jordan 2005; B. A. Pearl Mutter and Olsson 2006). Further the spectrum of the background noise has to be estimated by subtracting the noise spectrum from the spectrum of the frame (M. Berouli et al 1979; Yasser Ghanbari and Md. Reza 2004). It has been observed that some of the noise remains in the spectrum when the value of noise is greater than its mean. At the same time, some of the speech spectrum gets removed when noise is greater than the actual value of noise. In the spectrum this produces negative values and have to be set to zero. The overall effect puts a noise in the output signal known as

This paper presents a novel method for representing speaker characteristics present in the speech signal, by the way of deemphasizing the linguistic content of the signal. Cepstral coefficients that are widely employed as features for automatic speaker recognition task, contain considerable speech information in addition to the speaker information, and hence do not highlight the latter. The proposed method is based on using the difference between all-pole spectra due to higher order and lower order of linear prediction analysis. Distribution of the feature vectors in the multi-dimensional feature space is captured by employing autoassociative neural network models. A speaker recognition system is developed using the proposed method of feature extraction, whose performance is evaluated against that of the system based on cepstral coefficients. The complementary nature of evidence due to the proposed feature is also examined, so as to improve the overall system performance.

The presence of acoustic noise in audio recordings is an ongoing issue that plagues many applications. This ambient background noise is difficult to reduce due to its unpredictable nature. Many single channel noise reduction techniques exist but are limited in that they may distort the desired speech signal due to overlapping spectral content of the speech and noise. It is therefore of interest to investigate the use of multichannel noise reduction algorithms to further attenuate noise while attempting to preserve the speech signal of interest. Specifically, this thesis looks to investigate the use of microphone arrays in conjunction with multichannel noise reduction algorithms to aid aiding in speaker identification. Recording a speaker in the presence of acoustic background noise ultimately limits the performance and confidence of speaker identification algorithms. In situations where it is impossible to control the noise environment where the speech sample is taken, noise reduction algorithms must be developed and applied to clean the speech signal in order to give speaker identification software a chance at a positive identification. Due to the limitations of single channel techniques, it is of interest to see if spatial information provided by microphone arrays can be exploited to aid in speaker identification. This thesis provides an exploration of several time domain multichannel noise reduction techniques including delay sum beamforming, multi-channel Wiener filtering, and Spatial-Temporal Prediction filtering. Each algorithm is prototyped and filter performance is evaluated using various simulations and experiments. A three-dimensional noise model is developed to simulate and compare the performance of the above methods and experimental results of three data collections are presented and analyzed. The algorithms are compared and recommendations are given for the use of each technique. Finally, ideas for future work are discussed to improve performance and implementation of these multichannel algorithms. Possible applications for this technology include audio surveillance, identity verification, video chatting, conference calling and sound source localization.

Abstract— Audio data carries rich information in the form of speech, music, and environmental sounds, but its raw waveform is often complex and high-dimensional, making direct analysis difficult. Neural network architectures have emerged as powerful tools for extracting meaningful representations from audio signals, enabling efficient analysis and interpretation. Recurrent Neural Networks (RNNs), and Transformer-based models— for learning robust and discriminative features from audio. By automatically capturing temporal, spectral, and contextual patterns, these architectures significantly improve performance in tasks such as speech recognition, speaker identification, music classification, and environmental sound detection. The findings highlight the potential of neural networks to replace traditional handcrafted features, thereby advancing the development of scalable, accurate, and realtime audio processing applications. The rapid growth of audio data across domains such as speech, music, healthcare, and environmental monitoring has created a strong need for effective methods to extract meaningful representations from complex audio signals. Traditional approaches rely on handcrafted features like MFCCs and spectrogram descriptors, which often fail to capture the full temporal and spectral dynamics present in raw audio. Keywords: Neural Networks, Deep Learning, Audio Representation Learning, Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Transformer Models, Feature Extraction, Speech Recognition, Speaker Identification, Sound Event Detection, Spectrogram Analysis, Audio Signal Processing

Automatic Speech and Speaker Recognition technology has growing demands in variety of voice operated devices. Although the input for all such systems is speech signal, the features useful for each application/task are different. Of the different speech sounds, vowel sounds spectrally well-defined and well represented by formants. Formants which represent resonances of vocal tract are the result of physiology of individual’s speech production mechanism as well as nature of speech (words) being spoken. In this way formants are features of speech as well as of speaker. In this paper significance of formants for speech and speaker recognition is explored through experimental analysis. Formant tracking and estimation is done using adaptive formant filter bank and single pole formant based filter. Twelve vowel sounds represented in ARPABET (Advanced Research Project Agency bet) form are used to estimate the first four formants. The analysis based on extracting and emphasizing speaker specific clues indicates that higher formants carry more speaker specific information than first (lower) formant.

Residual networks for text-independent speaker identification: Unleashing the power of residual learning