Publications

Refine Results

(Filters Applied) Clear All

Advances in cross-lingual and cross-source audio-visual speaker recognition: The JHU-MIT system for NIST SRE21

Summary

We present a condensed description of the joint effort of JHUCLSP/HLTCOE, MIT-LL and AGH for NIST SRE21. NIST SRE21 consisted of speaker detection over multilingual conversational telephone speech (CTS) and audio from video (AfV). Besides the regular audio track, the evaluation also contains visual (face recognition) and multi-modal tracks. This evaluation exposes new challenges, including cross-source–i.e., CTS vs. AfV– and cross-language trials. Each speaker can speak two or three languages among English, Mandarin and Cantonese. For the audio track, we evaluated embeddings based on Res2Net and ECAPA-TDNN, where the former performed the best. We used PLDA based back-ends trained on previous SRE and VoxCeleb and adapted to a subset of Mandarin/Cantonese speakers. Some novel contributions of this submission are: the use of neural bandwidth extension (BWE) to reduce the mismatch between the AFV and CTS conditions; and invariant representation learning (IRL) to make the embeddings from a given speaker invariant to language. Res2Net with neural BWE was the best monolithic system. We used a pre-trained RetinaFace face detector and ArcFace embeddings for the visual track, following our NIST SRE19 work. We also included a new system using a deep pyramid single shot face detector and face embeddings trained on Crystal loss and probabilistic triplet loss, which performed the best. The number of face embeddings in the test video was reduced by agglomerative clustering or weighting the embedding based on the face detection confidence. Cosine scoring was used to compare embeddings. For the multi-modal track, we just added the calibrated likelihood ratios of the audio and visual conditions, assuming independence between modalities. The multi-modal fusion improved Cprimary by 72% w.r.t. audio.
READ LESS

Summary

We present a condensed description of the joint effort of JHUCLSP/HLTCOE, MIT-LL and AGH for NIST SRE21. NIST SRE21 consisted of speaker detection over multilingual conversational telephone speech (CTS) and audio from video (AfV). Besides the regular audio track, the evaluation also contains visual (face recognition) and multi-modal tracks. This...

READ MORE

Advances in speaker recognition for multilingual conversational telephone speech: the JHU-MIT system for NIST SRE20 CTS challenge

Published in:
Speaker and Language Recognition Workshop, Odyssey 2022, pp. 338-345.

Summary

We present a condensed description of the joint effort of JHUCLSP/HLTCOE and MIT-LL for NIST SRE20. NIST SRE20 CTS consisted of multilingual conversational telephone speech. The set of languages included in the evaluation was not provided, encouraging the participants to develop systems robust to any language. We evaluated x-vector architectures based on ResNet, squeeze-excitation ResNets, Transformers and EfficientNets. Though squeeze-excitation ResNets and EfficientNets provide superior performance in in-domain tasks like VoxCeleb, regular ResNet34 was more robust in the challenge scenario. On the contrary, squeeze-excitation networks over-fitted to the training data, mostly in English. We also proposed a novel PLDA mixture and k-NN PLDA back-ends to handle the multilingual trials. The former clusters the x-vector space expecting that each cluster will correspond to a language family. The latter trains a PLDA model adapted to each enrollment speaker using the nearest speakers–i.e., those with similar language/channel. The k-NN back-end improved Act. Cprimary (Cp) by 68% in SRE16-19 and 22% in SRE20 Progress w.r.t. a single adapted PLDA back-end. Our best single system achieved Act. Cp=0.110 in SRE20 progress. Meanwhile, our best fusion obtained Act. Cp=0.110 in the progress–8% better than single– and Cp=0.087 in the eval set.
READ LESS

Summary

We present a condensed description of the joint effort of JHUCLSP/HLTCOE and MIT-LL for NIST SRE20. NIST SRE20 CTS consisted of multilingual conversational telephone speech. The set of languages included in the evaluation was not provided, encouraging the participants to develop systems robust to any language. We evaluated x-vector architectures...

READ MORE

Quantifying bias in face verification system

Summary

Machine learning models perform face verification (FV) for a variety of highly consequential applications, such as biometric authentication, face identification, and surveillance. Many state-of-the-art FV systems suffer from unequal performance across demographic groups, which is commonly overlooked by evaluation measures that do not assess population-specific performance. Deployed systems with bias may result in serious harm against individuals or groups who experience underperformance. We explore several fairness definitions and metrics, attempting to quantify bias in Google’s FaceNet model. In addition to statistical fairness metrics, we analyze clustered face embeddings produced by the FV model. We link well-clustered embeddings (well-defined, dense clusters) for a demographic group to biased model performance against that group. We present the intuition that FV systems underperform on protected demographic groups because they are less sensitive to differences between features within those groups, as evidenced by clustered embeddings. We show how this performance discrepancy results from a combination of representation and aggregation bias.
READ LESS

Summary

Machine learning models perform face verification (FV) for a variety of highly consequential applications, such as biometric authentication, face identification, and surveillance. Many state-of-the-art FV systems suffer from unequal performance across demographic groups, which is commonly overlooked by evaluation measures that do not assess population-specific performance. Deployed systems with bias...

READ MORE

Bayesian estimation of PLDA in the presence of noisy training labels, with applications to speaker verification

Published in:
IEEE/ACM Trans. Audio, Speech, Language Process., Vol. 30, 2022, pp. 414-28.

Summary

This paper presents a Bayesian framework for estimating a Probabilistic Linear Discriminant Analysis (PLDA) model in the presence of noisy labels. True class labels are interpreted as latent random variables, which are transmitted through a noisy channel, and received as observed speaker labels. The labeling process is modeled as a Discrete Memoryless Channel (DMC). PLDA hyperparameters are interpreted as random variables, and their joint posterior distribution is derived using meanfield Variational Bayes, allowing maximum a posteriori (MAP) estimates of the PLDA model parameters to be determined. The proposed solution, referred to as VB-MAP, is presented as a general framework, but is studied in the context of speaker verification, and a variety of use cases are discussed. Specifically, VB-MAP can be used for PLDA estimation with unreliable labels, unsupervised PLDA estimation, and to infer the reliability of a PLDA training set. Experimental results show the proposed approach to provide significant performance improvements on a variety of NIST Speaker Recognition Evaluation (SRE) tasks, both for data sets with simulated mislabels, and for data sets with naturally occurring missing or unreliable labels.
READ LESS

Summary

This paper presents a Bayesian framework for estimating a Probabilistic Linear Discriminant Analysis (PLDA) model in the presence of noisy labels. True class labels are interpreted as latent random variables, which are transmitted through a noisy channel, and received as observed speaker labels. The labeling process is modeled as a...

READ MORE

Tools and practices for responsible AI engineering

Summary

Responsible Artificial Intelligence (AI)—the practice of developing, evaluating, and maintaining accurate AI systems that also exhibit essential properties such as robustness and explainability—represents a multifaceted challenge that often stretches standard machine learning tooling, frameworks, and testing methods beyond their limits. In this paper, we present two new software libraries—hydra-zen and the rAI-toolbox—that address critical needs for responsible AI engineering. hydra-zen dramatically simplifies the process of making complex AI applications configurable, and their behaviors reproducible. The rAI-toolbox is designed to enable methods for evaluating and enhancing the robustness of AI-models in a way that is scalable and that composes naturally with other popular ML frameworks. We describe the design principles and methodologies that make these tools effective, including the use of property-based testing to bolster the reliability of the tools themselves. Finally, we demonstrate the composability and flexibility of the tools by showing how various use cases from adversarial robustness and explainable AI can be concisely implemented with familiar APIs.
READ LESS

Summary

Responsible Artificial Intelligence (AI)—the practice of developing, evaluating, and maintaining accurate AI systems that also exhibit essential properties such as robustness and explainability—represents a multifaceted challenge that often stretches standard machine learning tooling, frameworks, and testing methods beyond their limits. In this paper, we present two new software libraries—hydra-zen and...

READ MORE

Adapting deep learning models to new meteorological contexts using transfer learning

Published in:
2021 IEEE International Conference on Big Data (Big Data), 2021, pp. 4169-4177, doi: 10.1109/BigData52589.2021.9671451.

Summary

Meteorological applications such as precipitation nowcasting, synthetic radar generation, statistical downscaling and others have benefited from deep learning (DL) approaches, however several challenges remain for widespread adaptation of these complex models in operational systems. One of these challenges is adequate generalizability; deep learning models trained from datasets collected in specific contexts should not be expected to perform as well when applied to different contexts required by large operational systems. One obvious mitigation for this is to collect massive amounts of training data that cover all expected meteorological contexts, however this is not only costly and difficult to manage, but is also not possible in many parts of the globe where certain sensing platforms are sparse. In this paper, we describe an application of transfer learning to perform domain transfer for deep learning models. We demonstrate a transfer learning algorithm called weight superposition to adapt a Convolutional Neural Network trained in a source context to a new target context. Weight superposition is a method for storing multiple models within a single set of parameters thus greatly simplifying model maintenance and training. This approach also addresses the issue of catastrophic forgetting where a model, once adapted to a new context, performs poorly in the original context. We apply weight superposition to the problem of synthetic weather radar generation and show that in scenarios where the target context has less data, a model adapted with weight superposition is better at maintaining performance when compared to simpler methods. Conversely, the simple adapted model performs better on the source context when the source and target contexts have comparable amounts of data.
READ LESS

Summary

Meteorological applications such as precipitation nowcasting, synthetic radar generation, statistical downscaling and others have benefited from deep learning (DL) approaches, however several challenges remain for widespread adaptation of these complex models in operational systems. One of these challenges is adequate generalizability; deep learning models trained from datasets collected in specific...

READ MORE

Unsupervised Bayesian adaptation of PLDA for speaker verification

Published in:
Interspeech, 30 August - 3 September 2021.

Summary

This paper presents a Bayesian framework for unsupervised domain adaptation of Probabilistic Linear Discriminant Analysis (PLDA). By interpreting class labels as latent random variables, Variational Bayes (VB) is used to derive a maximum a posterior (MAP) solution of the adapted PLDA model when labels are missing, referred to as VB-MAP. The VB solution iteratively infers class labels and updates PLDA hyperparameters, offering a systematic framework for dealing with unlabeled data. While presented as a general solution, this paper includes experimental results for domain adaptation in speaker verification. VBMAP estimation is applied to the 2016 and 2018 NIST Speaker Recognition Evaluations (SREs), both of which included small and unlabeled in-domain data sets, and is shown to provide performance improvements over a variety of state-of-the-art domain adaptation methods. Additionally, VB-MAP estimation is used to train a fully unsupervised PLDA model, suffering only minor performance degradation relative to conventional supervised training, offering promise for training PLDA models when no relevant labeled data exists.
READ LESS

Summary

This paper presents a Bayesian framework for unsupervised domain adaptation of Probabilistic Linear Discriminant Analysis (PLDA). By interpreting class labels as latent random variables, Variational Bayes (VB) is used to derive a maximum a posterior (MAP) solution of the adapted PLDA model when labels are missing, referred to as VB-MAP...

READ MORE

PATHATTACK: attacking shortest paths in complex networks

Summary

Shortest paths in complex networks play key roles in many applications. Examples include routing packets in a computer network, routing traffic on a transportation network, and inferring semantic distances between concepts on the World Wide Web. An adversary with the capability to perturb the graph might make the shortest path between two nodes route traffic through advantageous portions of the graph (e.g., a toll road he owns). In this paper, we introduce the Force Path Cut problem, in which there is a specific route the adversary wants to promote by removing a minimum number of edges in the graph. We show that Force Path Cut is NP-complete, but also that it can be recast as an instance of the Weighted Set Cover problem, enabling the use of approximation algorithms. The size of the universe for the set cover problem is potentially factorial in the number of nodes. To overcome this hurdle, we propose the PATHATTACK algorithm, which via constraint generation considers only a small subset of paths|at most 5% of the number of edges in 99% of our experiments. Across a diverse set of synthetic and real networks, the linear programming formulation of Weighted Set Cover yields the optimal solution in over 98% of cases. We also demonstrate a time/cost tradeoff using two approximation algorithms and greedy baseline methods. This work provides a foundation for addressing similar problems and expands the area of adversarial graph mining beyond recent work on node classification and embedding.
READ LESS

Summary

Shortest paths in complex networks play key roles in many applications. Examples include routing packets in a computer network, routing traffic on a transportation network, and inferring semantic distances between concepts on the World Wide Web. An adversary with the capability to perturb the graph might make the shortest path...

READ MORE

Combating Misinformation: HLT Highlights from MIT Lincoln Laboratory

Published in:
Human Language Technology Conference (HLTCon), 16-18 March 2021.

Summary

Dr. Joseph Campbell shares several human language technologies highlights from MIT Lincoln Laboratory. These include key enabling technologies in combating misinformation to link personas, analyze content, and understand human networks. Developing operationally relevant technologies requires access to corresponding data with meaningful evaluations, as Dr. Douglas Reynolds presented in his keynote. As Dr. Danelle Shah discussed in her keynote, it’s crucial to develop these technologies to operate at deeper levels than the surface. Producing reliable information from the fusion of missing and inherently unreliable information channels is paramount. Furthermore, the dynamic misinformation environment and the coevolution of allied methods with adversarial methods represent additional challenges
READ LESS

Summary

Dr. Joseph Campbell shares several human language technologies highlights from MIT Lincoln Laboratory. These include key enabling technologies in combating misinformation to link personas, analyze content, and understand human networks. Developing operationally relevant technologies requires access to corresponding data with meaningful evaluations, as Dr. Douglas Reynolds presented in his keynote...

READ MORE

Combating Misinformation: What HLT Can (and Can't) Do When Words Don't Say What They Mean

Author:
Published in:
Human Language Technology Conference (HLTCon), 16-18 March 2021.

Summary

Misinformation, disinformation, and “fake news” have been used as a means of influence for millennia, but the proliferation of the internet and social media in the 21st century has enabled nefarious campaigns to achieve unprecedented scale, speed, precision, and effectiveness. In the past few years, there has been significant recognition of the threats posed by malign influence operations to geopolitical relations, democratic institutions and processes, public health and safety, and more. At the same time, the digitization of communication offers tremendous opportunities for human language technologies (HLT) to observe, interpret, and understand this publicly available content. The ability to infer intent and impact, however, remains much more elusive.
READ LESS

Summary

Misinformation, disinformation, and “fake news” have been used as a means of influence for millennia, but the proliferation of the internet and social media in the 21st century has enabled nefarious campaigns to achieve unprecedented scale, speed, precision, and effectiveness. In the past few years, there has been significant recognition...

READ MORE