In this paper we explore speaker identification using electroencephalography (EEG) signals. The performance of speaker identification systems degrades in presence of background noise, this paper demonstrates that EEG features can be used to enhance the performance of speaker identification systems operating in presence and absence of background noise. The paper further demonstrates that in presence of high background noise, speaker identification system using only EEG features as input demonstrates better performance than the system using only acoustic features as input.
In this paper we demonstrate that performance of a speaker verification system can be improved by concatenating electroencephalography (EEG) signal features with speech signal features or only using EEG signal features. We use state-of-the-art end-to-end deep learning model for performing speaker verification and we demonstrate our results for noisy speech. Our results indicate that EEG signals can improve the robustness of speaker verification systems, especially in noiser environment.
EEG-Derived Voice Signature for Attended Speaker Detection
National University of Singapore（新加坡国立大学）；University of New South Wales（新南威尔士大学）；Shenzhen Research Institute of Big Data（深圳大数据研究院）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Kriston AI Lab（科斯顿人工智能实验室）；University of Bremen（不来梅大学）
Human brain performs remarkably well in segregating a particular speaker from interfering ones in a multi-speaker scenario. It has been recently shown that we can quantitatively evaluate the segregation capability by modelling the relationship between the speech signals present in an auditory scene and the cortical signals of the listener measured using electroencephalography (EEG). This has opened up avenues to integrate neuro-feedback into hearing aids whereby the device can infer user's attention and enhance the attended speaker. Commonly used algorithms to infer the auditory attention are …
Improved Feature Extraction Network for Neuro-Oriented Target Speaker Extraction
School of Computer Science and Technology, Anhui University（安徽大学计算机科学与技术学院）；National University of Singapore（新加坡国立大学）；University of Science and Technology of China (USTC)（中国科学技术大学）
EEG-informed attended speaker extraction from recorded speech mixtures with application in neuro-steered hearing prostheses
ESAT Laboratory of KU Leuven（KU莱顿大学ESAT实验室）；KU Leuven（KU莱顿大学）；Department of Electrical Engineering (ESAT)（电气工程系（ESAT））；Department of Neurosciences（神经科学系）
This paper presents a study on discriminative artificial neural network classifiers in the context of open-set speaker identification. Both 2-class and multi-class architectures are tested against the conventional Gaussian mixture model based classifier on enrolled speaker sets of different sizes. The performance evaluation shows that the multi-class neural network system has superior performance for large population sizes.
Speaker identification systems in a real-world scenario are tasked to identify a speaker amongst a set of enrolled speakers given just a few samples for each enrolled speaker. This paper demonstrates the effectiveness of meta-learning and relation networks for this use case. We propose improved relation networks for speaker verification and few-shot (unseen) speaker identification. The use of relation networks facilitates joint training of the frontend speaker encoder and the backend model. Inspired by the use of prototypical networks in speaker verification and to increase the discriminabilit…
Even human intelligence system fails to offer 100% accuracy in identifying speeches from a specific individual. Machine intelligence is trying to mimic humans in speaker identification problems through various approaches to speech feature extraction and speech modeling techniques. This paper presents a text-independent speaker identification system that employs Mel Frequency Cepstral Coefficients (MFCC) for feature extraction and k-Nearest Neighbor (kNN) for classification. The maximum cross-validation accuracy obtained was 60%. This will be improved upon in subsequent research.
Speaker identification models are vulnerable to carefully designed adversarial perturbations of their input signals that induce misclassification. In this work, we propose a white-box steganography-inspired adversarial attack that generates imperceptible adversarial perturbations against a speaker identification model. Our approach, FoolHD, uses a Gated Convolutional Autoencoder that operates in the DCT domain and is trained with a multi-objective loss function, in order to generate and conceal the adversarial perturbation within the original audio files. In addition to hindering speaker ident…
We propose an end-to-end speaker-attributed automatic speech recognition model that unifies speaker counting, speech recognition, and speaker identification on monaural overlapped speech. Our model is built on serialized output training (SOT) with attention-based encoder-decoder, a recently proposed method for recognizing overlapped speech comprising an arbitrary number of speakers. We extend SOT by introducing a speaker inventory as an auxiliary input to produce speaker labels as well as multi-speaker transcriptions. All model parameters are optimized by speaker-attributed maximum mutual info…
Adversarial attack approaches to speaker identification either need high computational cost or are not very effective, to our knowledge. To address this issue, in this paper, we propose a novel generation-network-based approach, called symmetric saliency-based encoder-decoder (SSED), to generate adversarial voice examples to speaker identification. It contains two novel components. First, it uses a novel saliency map decoder to learn the importance of speech samples to the decision of a targeted speaker identification system, so as to make the attacker focus on generating artificial noise to t…
We propose a novel algorithm for adaptive blind audio source extraction. The proposed method is based on independent vector analysis and utilizes the auxiliary function optimization to achieve high convergence speed. The algorithm is partially supervised by a pilot signal related to the source of interest (SOI), which ensures that the method correctly extracts the utterance of the desired speaker. The pilot is based on the identification of a dominant speaker in the mixture using x-vectors. The properties of the x-vectors computed in the presence of cross-talk are experimentally analyzed. The …
We propose a new feature, namely, pitchsynchronous discrete cosine transform (PS-DCT), for the task of speaker identification. These features are obtained directly from the voiced segments of the speech signal, without any preemphasis or windowing. The feature vectors are vector quantized, to create one separate codebook for each speaker during training. The performance of the PS-DCT features is shown to be good, and hence it can be used to supplement other features for the speaker identification task. Speaker identification is also performed using Mel-frequency cepstral coefficient (MFCC) fea…
TecnoCampus Mataró-Maresme（特诺坎普斯马塔罗-马雷斯梅）；Università Politecnica delle Marche（马尔凯理工大学）
The present work is based on the COST Action IC1206 for De-identification in multimedia content. It was performed to test four algorithms of voice modifications on a speech gender recognizer to find the degree of modification of pitch when the speech recognizer have the probability of success equal to the probability of failure. The purpose of this analysis is to assess the intensity of the speech tone modification, the quality, the reversibility and not-reversibility of the changes made. Keywords DeIdentification; Speech Algorithms
Escola Universitària Politècnica de Mataró (UPC)（马塔罗理工学院（加泰罗尼亚理工大学））
In this paper we discuss the relevance of bandwidth extension for speaker identification tasks. Mainly we want to study if it is possible to recognize voices that have been bandwith extended. For this purpose, we created two different databases (microphonic and ISDN) of speech signals that were bandwidth extended from telephone bandwidth ([300, 3400] Hz) to full bandwidth ([100, 8000] Hz). We have evaluated different parameterizations, and we have found that the MELCEPST parameterization can take advantage of the bandwidth extension algorithms in several situations.
Deep embedding based text-independent speaker verification has demonstrated superior performance to traditional methods in many challenging scenarios. Its loss functions can be generally categorized into two classes, i.e., verification and identification. The verification loss functions match the pipeline of speaker verification, but their implementations are difficult. Thus, most state-of-the-art deep embedding methods use the identification loss functions with softmax output units or their variants. In this paper, we propose a verification loss function, named the maximization of partial are…
Pattern Recognition and Intelligent System Lab., Beijing University of Posts and Telecommunications（北京邮电大学模式识别与智能系统实验室，北京邮电大学）
A novel text-independent speaker identification (SI) method is proposed. This method uses the Mel-frequency Cepstral coefficients (MFCCs) and the dynamic information among adjacent frames as feature sets to capture speaker's characteristics. In order to utilize dynamic information, we design super-MFCCs features by cascading three neighboring MFCCs frames together. The probability density function (PDF) of these super-MFCCs features is estimated by the recently proposed histogram transform~(HT) method, which generates more training data by random transforms to realize the histogram PDF estimat…
Indian Institute of Technology Guwahati（印度理工学院古瓦哈提分校）；National University of Singapore（新加坡国立大学）
Emotional state of a speaker is found to have significant effect in speech production, which can deviate speech from that arising from neutral state. This makes identifying speakers with different emotions a challenging task as generally the speaker models are trained using neutral speech. In this work, we propose to overcome this problem by creation of emotion invariant speaker embedding. We learn an extractor network that maps the test embeddings with different emotions obtained using i-vector based system to an emotion invariant space. The resultant test embeddings thus become emotion invar…
Escola Universitària Politècnica de Mataró（马塔罗理工大学学院）；Universitat Politècnica de Catalunya（加泰罗尼亚理工大学）
In this paper we evaluate the relevance of the model size for speaker identification. We show that it is possible to improve the identification rates if a different model size is used for each speaker. We also present some criteria for selecting the model size, and a new algorithm that outperforms the classical system with a fixed model size.
Worcester Polytechnic Institute (WPI)（伍斯特理工学院（WPI））
We propose a new method for speaker diarization that can handle overlapping speech with 2+ people. Our method is based on compositional embeddings [1]: Like standard speaker embedding methods such as x-vector [2], compositional embedding models contain a function f that separates speech from different speakers. In addition, they include a composition function g to compute set-union operations in the embedding space so as to infer the set of speakers within the input audio. In an experiment on multi-person speaker identification using synthesized LibriSpeech data, the proposed method outperform…
We propose an approach for training speaker identification models in a weakly supervised manner. We concentrate on the setting where the training data consists of a set of audio recordings and the speaker annotation is provided only at the recording level. The method uses speaker diarization to find unique speakers in each recording, and i-vectors to project the speech of each speaker to a fixed-dimensional vector. A neural network is then trained to map i-vectors to speakers, using a special objective function that allows to optimize the model using recording-level speaker labels. We report e…
University of Rajshahi（拉杰沙希大学）；International Islamic University Chittagong（吉大港国际伊斯兰大学）
Speech recognition and speaker identification are important for authentication and verification in security purpose, but they are difficult to achieve. Speaker identification methods can be divided into text-independent and text-dependent. This paper presents a technique of text-dependent speaker identification using MFCC-domain support vector machine (SVM). In this work, melfrequency cepstrum coefficients (MFCCs) and their statistical distribution properties are used as features, which will be inputs to the neural network. This work firstly used sequential minimum optimization (SMO) learning …
An efficient, and intuitive algorithm is presented for the identification of speakers from a long dataset (like YouTube long discussion, Cocktail party recorded audio or video).The goal of automatic speaker identification is to identify the number of different speakers and prepare a model for that speaker by extraction, characterization and speaker-specific information contained in the speech signal. It has many diverse application specially in the field of Surveillance, Immigrations at Airport, cyber security, transcription in multi-source of similar sound source, where it is difficult to ass…
Singapore University of Technology and Design（新加坡科技设计大学）；Institute of High Performance Computing, A*STAR（A*STAR高性能计算研究所）
We present an approach to tackle the speaker recognition problem using Triplet Neural Networks. Currently, the $i$-vector representation with probabilistic linear discriminant analysis (PLDA) is the most commonly used technique to solve this problem, due to high classification accuracy with a relatively short computation time. In this paper, we explore a neural network approach, namely Triplet Neural Networks (TNNs), to built a latent space for different classifiers to solve the Multi-Target Speaker Detection and Identification Challenge Evaluation 2018 (MCE 2018) dataset. This training set co…
Speaker identification in the household scenario (e.g., for smart speakers) is typically based on only a few enrollment utterances but a much larger set of unlabeled data, suggesting semisupervised learning to improve speaker profiles. We propose a graph-based semi-supervised learning approach for speaker identification in the household scenario, to leverage the unlabeled speech samples. In contrast to most of the works in speaker recognition that focus on speaker-discriminative embeddings, this work focuses on speaker label inference (scoring). Given a pre-trained embedding extractor, graph-b…
This paper examines the speaker identification potential of breath sounds in continuous speech. Speech is largely produced during exhalation. In order to replenish air in the lungs, speakers must periodically inhale. When inhalation occurs in the midst of continuous speech, it is generally through the mouth. Intra-speech breathing behavior has been the subject of much study, including the patterns, cadence, and variations in energy levels. However, an often ignored characteristic is the {\em sound} produced during the inhalation phase of this cycle. Intra-speech inhalation is rapid and energet…
Speech 'in-the-wild' is a handicap for speaker recognition systems due to the variability induced by real-life conditions, such as environmental noise and the emotional state of the speaker. Taking advantage of the principles of representation learning, we aim to design a recurrent denoising autoencoder that extracts robust speaker embeddings from noisy spectrograms to perform speaker identification. The end-to-end proposed architecture uses a feedback loop to encode information regarding the speaker into low-dimensional representations extracted by a spectrogram denoising autoencoder. We empl…
M3ANet: Multi-scale and Multi-Modal Alignment Network for Brain-Assisted Target Speaker Extraction
School of Computer Science and Technology, Anhui University（安徽大学计算机科学与技术学院）；Alibaba group（阿里巴巴集团）；Department of Automation, Tsinghua University（清华大学自动化系）；Beijing National Research Center for lnformation Science and Technology, Tsinghua University（清华大学信息科学与技术国家研究中心）
Speaker identification typically involves three stages. First, a front-end speaker embedding model is trained to embed utterance and speaker profiles. Second, a scoring function is applied between a runtime utterance and each speaker profile. Finally, the speaker is identified using nearest neighbor according to the scoring metric. To better distinguish speakers sharing a device within the same household, we propose a household-adapted nonlinear mapping to a low dimensional space to complement the global scoring metric. The combined scoring function is optimized on labeled or pseudo-labeled sp…
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% ba…
Conversational emotion recognition (CER) has attracted increasing interests in the natural language processing (NLP) community. Different from the vanilla emotion recognition, effective speaker-sensitive utterance representation is one major challenge for CER. In this paper, we exploit speaker identification (SI) as an auxiliary task to enhance the utterance representation in conversations. By this method, we can learn better speaker-aware contextual representations from the additional SI corpus. Experiments on two benchmark datasets demonstrate that the proposed architecture is highly effecti…
Instituto de Investigaciones en Matemáticas Aplicadas y en Sistemas (IIMAS), Universidad Nacional Autónoma de México (UNAM)（墨西哥国立自治大学应用数学与系统研究所（IIMAS））
Verifying if two audio segments belong to the same speaker has been recently put forward as a flexible way to carry out speaker identification, since it does not require to be re-trained when new speakers appear on the auditory scene. Although many of the current techniques have achieved high performances, they require a considerably high amount of memory, and a specific minimum length for their input audio segments. These requirements limit the applicability of these techniques in scenarios such as service robots, internet of things and virtual assistants, where computational resources are li…
South China University of Technology（华南理工大学）；Guangdong Polytechnic of Science and Technology（广东科学技术职业学院）
Although few-shot learning has attracted much attention from the fields of image and audio classification, few efforts have been made on few-shot speaker identification. In the task of few-shot learning, overfitting is a tough problem mainly due to the mismatch between training and testing conditions. In this paper, we propose a few-shot speaker identification method which can alleviate the overfitting problem. In the proposed method, the model of a depthwise separable convolutional network with channel attention is trained with a prototypical loss function. Experimental datasets are extracted…
University of California, Berkeley（加州大学伯克利分校）；Research Institutes of Sweden（瑞典研究院）
We introduce a method to identify speakers by computing with high-dimensional random vectors. Its strengths are simplicity and speed. With only 1.02k active parameters and a 128-minute pass through the training data we achieve Top-1 and Top-5 scores of 31% and 52% on the VoxCeleb1 dataset of 1,251 speakers. This is in contrast to CNN models requiring several million parameters and orders of magnitude higher computational complexity for only a 2$\times$ gain in discriminative power as measured in mutual information. An additional 92 seconds of training with Generalized Learning Vector Quantizat…
This research is dedicated to improving text-independent Emirati-accented speaker identification performance in stressful talking conditions using three distinct classifiers: First-Order Hidden Markov Models (HMM1s), Second-Order Hidden Markov Models (HMM2s), and Third-Order Hidden Markov Models (HMM3s). The database that has been used in this work was collected from 25 per gender Emirati native speakers uttering eight widespread Emirati sentences in each of neutral, shouted, slow, loud, soft, and fast talking conditions. The extracted features of the captured database are called Mel-Frequency…
Enhancing Open-Set Speaker Identification through Rapid Tuning with Speaker Reciprocal Points and Negative Sample
School of Communication and Information Engineering, Shanghai University（上海大学通信与信息工程学院）；School of Communication（通信学院）；Information Engineering, Shanghai University（上海大学信息工程学院）
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), …
Speaker Identification process is to identify a particular vocal cord from a set of existing speakers. In the speaker identification processes, unknown speaker voice sample targets each of the existing speakers present in the system and gives a predication. The predication may be more than one existing known speaker voice and is very close to the unknown speaker voice. The model is a Gaussian mixture model built by the extracted acoustic feature vectors from voice. The i-vector based dimension compression mapping function of the channel depended speaker, and super vector give better predicted …
National Engineering School of Tunis（突尼斯国家工程学院）；Université Tunis El Manar（突尼斯埃尔马纳尔大学）
Several speaker identification systems are giving good performance with clean speech but are affected by the degradations introduced by noisy audio conditions. To deal with this problem, we investigate the use of complementary information at different levels for computing a combined match score for the unknown speaker. In this work, we observe the effect of two supervised machine learning approaches including support vectors machines (SVM) and naïve bayes (NB). We define two feature vector sets based on mel frequency cepstral coefficients (MFCC) and relative spectral perceptual linear predicti…
It is well known that speaker identification performs extremely well in the neutral talking environments; however, the identification performance is declined sharply in the shouted talking environments. This work aims at proposing, implementing and testing a new approach to enhance the declined performance in the shouted talking environments. The new proposed approach is based on gender-dependent speaker identification using Suprasegmental Hidden Markov Models (SPHMMs) as classifiers. This proposed approach has been tested on two different and separate speech databases: our collected database …
One of the major parts of the voice recognition field is the choice of acoustic features which have to be robust against the variability of the speech signal, mismatched conditions, and noisy environments. Thus, different speech feature extraction techniques have been developed. In this paper, we investigate the robustness of several front-end techniques in Arabic speaker identification. We evaluate five different features in babble, factory and subway conditions at the various signal to noise ratios (SNR). The obtained results showed that two of the auditory feature i.e. gammatone frequency c…
Speaker identification in noisy audio recordings, specifically those from collaborative learning environments, can be extremely challenging. There is a need to identify individual students talking in small groups from other students talking at the same time. To solve the problem, we assume the use of a single microphone per student group without any access to previous large datasets for training. This dissertation proposes a method of speaker identification using cross-correlation patterns associated to an array of virtual microphones, centered around the physical microphone. The virtual micro…
Identifying multiple speakers without knowing where a speaker's voice is in a recording is a challenging task. In this paper, a hierarchical attention network is proposed to solve a weakly labelled speaker identification problem. The use of a hierarchical structure, consisting of a frame-level encoder and a segment-level encoder, aims to learn speaker related information locally and globally. Speech streams are segmented into fragments. The frame-level encoder with attention learns features and highlights the target related frames locally, and output a fragment based embedding. The segment-lev…
Escola Universitària Politècnica de Mataró（马塔罗理工学院）
In this paper we propose a new parameterization algorithm based on nonlinear prediction, which is an extension of the classical LPC parameters. The parameters performances are estimated by two different methods: the Arithmetic-Harmonic Sphericity (AHS) and the Auto-Regressive Vector Model (ARVM). Two different methods are proposed for the parameterization based on the Neural Predictive Coding (NPC): classical neural networks initialization and linear initialization. We applied these two parameters to speaker identification. The fist parameters obtained smaller rates. We show for the first para…
University of Sharjah（沙迦大学）；The University of Tokyo（东京大学）
This work aims at intensifying text-independent speaker identification performance in real application situations such as noisy and emotional talking conditions. This is achieved by incorporating two different modules: a Computational Auditory Scene Analysis CASA based pre-processing module for noise reduction and cascaded Gaussian Mixture Model Convolutional Neural Network GMM-CNN classifier for speaker identification followed by emotion recognition. This research proposes and evaluates a novel algorithm to improve the accuracy of speaker identification in emotional and highly-noise susceptib…
University of Sharjah（沙迦大学）；Prince Sattam bin Abdulaziz University（萨塔姆·本·阿卜杜勒-阿齐兹亲王大学）；Bolu Abant Izzet Baysal University（博卢阿班特伊泽特贝萨尔大学）
Speaker recognition systems are widely used in various applications to identify a person by their voice; however, the high degree of variability in speech signals makes this a challenging task. Dealing with emotional variations is very difficult because emotions alter the voice characteristics of a person; thus, the acoustic features differ from those used to train models in a neutral environment. Therefore, speaker recognition models trained on neutral speech fail to correctly identify speakers under emotional stress. Although considerable advancements in speaker identification have been made…
This work is devoted to capturing Emirati-accented speech database (Arabic United Arab Emirates database) in each of neutral and shouted talking environments in order to study and enhance text-independent Emirati-accented speaker identification performance in shouted environment based on each of First-Order Circular Suprasegmental Hidden Markov Models (CSPHMM1s), Second-Order Circular Suprasegmental Hidden Markov Models (CSPHMM2s), and Third-Order Circular Suprasegmental Hidden Markov Models (CSPHMM3s) as classifiers. In this research, our database was collected from fifty Emirati native speak…
In this paper, second-order hidden Markov model (HMM2) has been used and implemented to improve the recognition performance of text-dependent speaker identification systems under neutral talking condition. Our results show that HMM2 improves the recognition performance under neutral talking condition compared to the first-order hidden Markov model (HMM1). The recognition performance has been improved by 9%.
Usable speech criteria are proposed to extract minimally corrupted speech for speaker identification (SID) in co-channel speech. In co-channel speech, either speaker can randomly appear as the stronger speaker or the weaker one at a time. Hence, the extracted usable segments are separated in time and need to be organized into speaker streams for SID. In this paper, we focus to organize extracted usable speech segment into a single stream for the same speaker by speaker assignment system. For this, we develop model-based speaker assignment method based on posterior probability and exhaustive se…
