Automated speaker recognition uses data processing to identify speakers by their voice. Today, automated speaker recognition is deployed on billions of smart devices and in services such as call centres. Despite their wide-scale deployment and known sources of bias in related domains like face recognition and natural language processing, bias in automated speaker recognition has not been studied systematically. We present an in-depth empirical and analytical study of bias in the machine learning development workflow of speaker verification, a voice biometric and core task in automated speaker recognition. Drawing on an established framework for understanding sources of harm in machine learning, we show that bias exists at every development stage in the well-known VoxCeleb Speaker Recognition Challenge, including data generation, model building, and implementation. Most affected are female speakers and non-US nationalities, who experience significant performance degradation. Leveraging the insights from our findings, we make practical recommendations for mitigating bias in automated speaker recognition, and outline future research directions.
Emergent Systems Laboratory, College of Information Science and Engineering, Ritsumeikan University（立命馆大学信息科学与工程学院新兴系统实验室）
This paper describes a new unsupervised machine learning method for simultaneous phoneme and word discovery from multiple speakers. Human infants can acquire knowledge of phonemes and words from interactions with his/her mother as well as with others surrounding him/her. From a computational perspective, phoneme and word discovery from multiple speakers is a more challenging problem than that from one speaker because the speech signals from different speakers exhibit different acoustic features. This paper proposes an unsupervised phoneme and word discovery method that simultaneously uses nonp…
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…
In forensic applications, it is very common that only small naturalistic datasets consisting of short utterances in complex or unknown acoustic environments are available. In this study, we propose a pipeline solution to improve speaker verification on a small actual forensic field dataset. By leveraging large-scale out-of-domain datasets, a knowledge distillation based objective function is proposed for teacher-student learning, which is applied for short utterance forensic speaker verification. The objective function collectively considers speaker classification loss, Kullback-Leibler diverg…
SAMOVAR, Telecom SudParis, Institut Polytechnique de Paris（SAMOVAR，巴黎电信学院，巴黎综合理工学院）；Newsbridge
Speaker diarization, or the task of finding "who spoke and when", is now used in almost every speech processing application. Nevertheless, its fairness has not yet been evaluated because there was no protocol to study its biases one by one. In this paper we propose a protocol and a scoring method designed to evaluate speaker diarization fairness. This protocol is applied on a large dataset of spoken utterances and report the performances of speaker diarization depending on the gender, the age, the accent of the speaker and the length of the spoken sentence. Some biases induced by the gender, o…
With the ubiquity of smart devices that use speaker recognition (SR) systems as a means of authenticating individuals and personalizing their services, fairness of SR systems has becomes an important point of focus. In this paper we study the notion of fairness in recent SR systems based on 3 popular and relevant definitions, namely Statistical Parity, Equalized Odds, and Equal Opportunity. We examine 5 popular neural architectures and 5 commonly used loss functions in training SR systems, while evaluating their fairness against gender and nationality groups. Our detailed experiments shed ligh…
Speaker embedding has been a fundamental feature for speaker-related tasks such as verification, clustering, and diarization. Traditionally, speaker embeddings are represented as fixed vectors in high-dimensional space. This could lead to biased estimations, especially when handling shorter utterances. In this paper we propose to represent a speaker utterance as "floating" vector whose state is indeterminate without knowing the context. The state of a speaker representation is jointly determined by itself, other speech from the same speaker, as well as other speakers it is being compared to. T…
The objective of this paper is speaker recognition under noisy and unconstrained conditions. We make two key contributions. First, we introduce a very large-scale audio-visual speaker recognition dataset collected from open-source media. Using a fully automated pipeline, we curate VoxCeleb2 which contains over a million utterances from over 6,000 speakers. This is several times larger than any publicly available speaker recognition dataset. Second, we develop and compare Convolutional Neural Network (CNN) models and training strategies that can effectively recognise identities from voice under…
Speaker diarization is one of the critical components of computational media intelligence as it enables a character-level analysis of story portrayals and media content understanding. Automated audio-based speaker diarization of entertainment media poses challenges due to the diverse acoustic conditions present in media content, be it background music, overlapping speakers, or sound effects. At the same time, speaking faces in the visual modality provide complementary information and not prone to the errors seen in the audio modality. In this paper, we address the problem of speaker diarizatio…
Facial recognition system is one of the major successes of Artificial intelligence and has been used a lot over the last years. But, images are not the only biometric present: audio is another possible biometric that can be used as an alternative to the existing recognition systems. However, the text-independent audio data is not always available for tasks like speaker verification and also no work has been done in the past for text-independent speaker verification assuming very little training data. Therefore, In this paper, we have proposed a novel method to verify the identity of the claime…
Despite the success of deep neural networks (DNNs) in enabling on-device voice assistants, increasing evidence of bias and discrimination in machine learning is raising the urgency of investigating the fairness of these systems. Speaker verification is a form of biometric identification that gives access to voice assistants. Due to a lack of fairness metrics and evaluation frameworks that are appropriate for testing the fairness of speaker verification components, little is known about how model performance varies across subgroups, and what factors influence performance variation. To tackle th…
I-vector based text-independent speaker verification (SV) systems often have poor performance with short utterances, as the biased phonetic distribution in a short utterance makes the extracted i-vector unreliable. This paper proposes an i-vector compensation method using a generative adversarial network (GAN), where its generator network is trained to generate a compensated i-vector from a short-utterance i-vector and its discriminator network is trained to determine whether an i-vector is generated by the generator or the one extracted from a long utterance. Additionally, we assign two other…
Recently, the attention mechanism such as squeeze-and-excitation module (SE) and convolutional block attention module (CBAM) has achieved great success in deep learning-based speaker verification system. This paper introduces an alternative effective yet simple one, i.e., simple attention module (SimAM), for speaker verification. The SimAM module is a plug-and-play module without extra modal parameters. In addition, we propose a noisy label detection method to iteratively filter out the data samples with a noisy label from the training data, considering that a large-scale dataset labeled with …
Most existing datasets for speaker identification contain samples obtained under quite constrained conditions, and are usually hand-annotated, hence limited in size. The goal of this paper is to generate a large scale text-independent speaker identification dataset collected 'in the wild'. We make two contributions. First, we propose a fully automated pipeline based on computer vision techniques to create the dataset from open-source media. Our pipeline involves obtaining videos from YouTube; performing active speaker verification using a two-stream synchronization Convolutional Neural Network…
Redwood Center for Theoretical Neuroscience（雷德伍德理论神经科学中心）；Helen Wills Neuroscience Institute（海伦·威尔斯神经科学研究所）；University of California, Berkeley（加州大学伯克利分校）
We present a model for separating a set of voices out of a sound mixture containing an unknown number of sources. Our Attentional Gating Network (AGN) uses a variable attentional context to specify which speakers in the mixture are of interest. The attentional context is specified by an embedding vector which modifies the processing of a neural network through an additive bias. Individual speaker embeddings are learned to separate a single speaker while superpositions of the individual speaker embeddings are used to separate sets of speakers. We first evaluate AGN on a traditional single speak…
Deep neural networks can learn complex and abstract representations, that are progressively obtained by combining simpler ones. A recent trend in speech and speaker recognition consists in discovering these representations starting from raw audio samples directly. Differently from standard hand-crafted features such as MFCCs or FBANK, the raw waveform can potentially help neural networks discover better and more customized representations. The high-dimensional raw inputs, however, can make training significantly more challenging. This paper summarizes our recent efforts to develop a neural arc…
Rajshahi University of Engineering & Technology (RUET)（拉杰沙希工程技术大学）
In this paper, an improved strategy for automated text dependent speaker identification system has been proposed in noisy environment. The identification process incorporates the Neuro- Genetic hybrid algorithm with cepstral based features. To remove the background noise from the source utterance, wiener filter has been used. Different speech pre-processing techniques such as start-end point detection algorithm, pre-emphasis filtering, frame blocking and windowing have been used to process the speech utterances. RCC, MFCC, MFCC, MFCC, LPC and LPCC have been used to extract the features. After …
Automated Pronunciation Evaluation for Korean Toddler Speech using Speech Diarization and Self-Supervised Learning
Artificial Intelligence Program University of San Francisco（人工智能项目 旧金山大学）；Department of Rehabilitation Medicine Seoul National University Bundang Hospital（康复医学系 首尔国立大学束草医院）
Speaker recognition performance in emotional talking environments is not as high as it is in neutral talking environments. This work focuses on proposing, implementing, and evaluating a new approach to enhance the performance in emotional talking environments. The new proposed approach is based on identifying the unknown speaker using both his/her gender and emotion cues. Both Hidden Markov Models (HMMs) and Suprasegmental Hidden Markov Models (SPHMMs) have been used as classifiers in this work. This approach has been tested on our collected emotional speech database which is composed of six e…
Center for Speech and Language Technologies, Tsinghua University（清华大学语音与语言技术研究中心）
Evaluation trials are used to probe performance of automatic speaker verification (ASV) systems. In spite of the clear importance and impact, evaluation trials have not been seriously treated in research and engineering practice. This paper firstly presents a theoretical analysis on evaluation trials and highlights potential bias with the most popular cross-pairing approach used in trials design. To interpret and settle this problem, we define the concept of trial config and C-P map derived from it. The C-P map measures the performance of an ASV system on various trial configs in a 2-dimension…
Isolating the voice of a specific person while filtering out other voices or background noises is challenging when video is shot in noisy environments. We propose audio-visual methods to isolate the voice of a single speaker and eliminate unrelated sounds. First, face motions captured in the video are used to estimate the speaker's voice, by passing the silent video frames through a video-to-speech neural network-based model. Then the speech predictions are applied as a filter on the noisy input audio. This approach avoids using mixtures of sounds in the learning process, as the number of such…
Automatic Speech Scoring (ASS) is the computer-assisted evaluation of a candidate's speaking proficiency in a language. ASS systems face many challenges like open grammar, variable pronunciations, and unstructured or semi-structured content. Recent deep learning approaches have shown some promise in this domain. However, most of these approaches focus on extracting features from a single audio, making them suffer from the lack of speaker-specific context required to model such a complex task. We propose a novel deep learning technique for non-native ASS, called speaker-conditioned hierarchical…
Naver Corporation（Naver公司）；University of Eastern Finland（东芬兰大学）；Korea Advanced Institute of Science and Technology（韩国科学技术院）；Carnegie Mellon University（卡内基梅隆大学）
The objective of this work is to develop a speaker recognition model to be used in diverse scenarios. We hypothesise that two components should be adequately configured to build such a model. First, adequate architecture would be required. We explore several recent state-of-the-art models, including ECAPA-TDNN and MFA-Conformer, as well as other baselines. Second, a massive amount of data would be required. We investigate several new training data configurations combining a few existing datasets. The most extensive configuration includes over 87k speakers' 10.22k hours of speech. Four evaluati…
National Institute of Informatics（国立信息学研究所）；The University of Edinburgh（爱丁堡大学）
Most neural-network based speaker-adaptive acoustic models for speech synthesis can be categorized into either layer-based or input-code approaches. Although both approaches have their own pros and cons, most existing works on speaker adaptation focus on improving one or the other. In this paper, after we first systematically overview the common principles of neural-network based speaker-adaptive models, we show that these approaches can be represented in a unified framework and can be generalized further. More specifically, we introduce the use of scaling and bias codes as generalized means f…
Benchmark Dataset Dynamics, Bias and Privacy Challenges in Voice Biometrics Research
University of Southern California（南加州大学）；Fraunhofer IIS（弗劳恩霍夫集成电路研究所）；University of Toronto（多伦多大学）；York University（约克大学）；Open North（北方开放组织）；Delft University of Technology（代尔夫特理工大学）
Toward Responsible ASR for African American English Speakers: A Scoping Review of Bias and Equity in Speech Technology
DePaul University（德保罗大学）；University of Washington（华盛顿大学）；Carnegie Mellon University（卡内基梅隆大学）；Cornell Tech University（康奈尔科技大学）；University of Maryland（马里兰大学）
Speak Your Mind: The Speech Continuation Task as a Probe of Voice-Based Model Bias
Department of Speech, Music and Hearing, KTH Royal Institute of Technology（语音、音乐与听觉系，皇家理工学院）；Univ. Grenoble Alpes, CNRS, Grenoble INP, GIPSA-lab（格勒诺布尔阿尔卑斯大学，法国国家科学研究中心，格勒诺布尔理工学院，GIPSA实验室）
National Institute of Informatics（国立信息学研究所）；SOKENDAI (The Graduate University for Advanced Studies)（综合研究大学院大学）；University of Edinburgh（爱丁堡大学）
By representing speaker characteristic as a single fixed-length vector extracted solely from speech, we can train a neural multi-speaker speech synthesis model by conditioning the model on those vectors. This model can also be adapted to unseen speakers regardless of whether the transcript of adaptation data is available or not. However, this setup restricts the speaker component to just a single bias vector, which in turn limits the performance of adaptation process. In this study, we propose a novel speech synthesis model, which can be adapted to unseen speakers by fine-tuning part of or all…
Structured Speaker-Deficiency Adaptation of Foundation Models for Dysarthric and Elderly Speech Recognition
The Chinese University of Hong Kong（香港中文大学）；Institute of Software, Chinese Academy of Sciences（中国科学院软件研究所）；National Research Council of Canada（加拿大国家研究委员会）
