Search

We propose an approach for simultaneous diarization and separation of meeting data. It consists of a complex Angular Central Gaussian Mixture Model (cACGMM) for speech source separation, and a von-Mises-Fisher Mixture Model (VMFMM) for diarization in a joint statistical framework. Through the integration, both spatial and spectral information are exploited for diarization and separation. We also develop a method for counting the number of active speakers in a segment of a meeting to support block-wise processing. While the total number of speakers in a meeting may be known, it is usually not known on a per-segment level. With the proposed speaker counting, joint diarization and source separation can be done segment-by-segment, and the permutation problem across segments is solved, thus allowing for block-online processing in the future. Experimental results on the LibriCSS meeting corpus show that the integrated approach outperforms a cascaded approach of diarization and speech enhancement in terms of WER, both on a per-segment and on a per-meeting level., Comment: Submitted to ICASSP2025

Diarization is a crucial component in meeting transcription systems to ease the challenges of speech enhancement and attribute the transcriptions to the correct speaker. Particularly in the presence of overlapping or noisy speech, these systems have problems reliably assigning the correct speaker labels, leading to a significant amount of speaker confusion errors. We propose to add segment-level speaker reassignment to address this issue. By revisiting, after speech enhancement, the speaker attribution for each segment, speaker confusion errors from the initial diarization stage are significantly reduced. Through experiments across different system configurations and datasets, we further demonstrate the effectiveness and applicability in various domains. Our results show that segment-level speaker reassignment successfully rectifies at least 40% of speaker confusion word errors, highlighting its potential for enhancing diarization accuracy in meeting transcription systems., Comment: Accepted for Interspeech 2024

Hyperbolic deep learning has become a growing research direction in computer vision due to the unique properties afforded by the alternate embedding space. The negative curvature and exponentially growing distance metric provide a natural framework for capturing hierarchical relationships between datapoints and allowing for finer separability between their embeddings. However, current hyperbolic learning approaches are still prone to overfitting, computationally expensive, and prone to instability, especially when attempting to learn the manifold curvature to adapt to tasks and different datasets. To address these issues, our paper presents a derivation for Riemannian AdamW that helps increase hyperbolic generalization ability. For improved stability, we introduce a novel fine-tunable hyperbolic scaling approach to constrain hyperbolic embeddings and reduce approximation errors. Using this along with our curvature-aware learning schema for Lorentzian Optimizers enables the combination of curvature and non-trivialized hyperbolic parameter learning. Our approach demonstrates consistent performance improvements across Computer Vision, EEG classification, and hierarchical metric learning tasks achieving state-of-the-art results in two domains and drastically reducing runtime.

We propose a modified teacher-student training for the extraction of frame-wise speaker embeddings that allows for an effective diarization of meeting scenarios containing partially overlapping speech. To this end, a geodesic distance loss is used that enforces the embeddings computed from regions with two active speakers to lie on the shortest path on a sphere between the points given by the d-vectors of each of the active speakers. Using those frame-wise speaker embeddings in clustering-based diarization outperforms segment-level clustering-based diarization systems such as VBx and Spectral Clustering. By extending our approach to a mixture-model-based diarization, the performance can be further improved, approaching the diarization error rates of diarization systems that use a dedicated overlap detection, and outperforming these systems when also employing an additional overlap detection., Comment: Accepted at ICASSP 2024

We propose a modular pipeline for the single-channel separation, recognition, and diarization of meeting-style recordings and evaluate it on the Libri-CSS dataset. Using a Continuous Speech Separation (CSS) system with a TF-GridNet separation architecture, followed by a speaker-agnostic speech recognizer, we achieve state-of-the-art recognition performance in terms of Optimal Reference Combination Word Error Rate (ORC WER). Then, a d-vector-based diarization module is employed to extract speaker embeddings from the enhanced signals and to assign the CSS outputs to the correct speaker. Here, we propose a syntactically informed diarization using sentence- and word-level boundaries of the ASR module to support speaker turn detection. This results in a state-of-the-art Concatenated minimum-Permutation Word Error Rate (cpWER) for the full meeting recognition pipeline., Comment: Accepted at HSCMA Sattelite Workshop at ICASSP 2024

We introduce a monaural neural speaker embeddings extractor that computes an embedding for each speaker present in a speech mixture. To allow for supervised training, a teacher-student approach is employed: the teacher computes the target embeddings from each speaker's utterance before the utterances are added to form the mixture, and the student embedding extractor is then tasked to reproduce those embeddings from the speech mixture at its input. The system much more reliably verifies the presence or absence of a given speaker in a mixture than a conventional speaker embedding extractor, and even exhibits comparable performance to a multi-channel approach that exploits spatial information for embedding extraction. Further, it is shown that a speaker embedding computed from a mixture can be used to check for the presence of that speaker in another mixture., Comment: Proceedings of INTERSPEECH

Using a Teacher-Student training approach we developed a speaker embedding extraction system that outputs embeddings at frame rate. Given this high temporal resolution and the fact that the student produces sensible speaker embeddings even for segments with speech overlap, the frame-wise embeddings serve as an appropriate representation of the input speech signal for an end-to-end neural meeting diarization (EEND) system. We show in experiments that this representation helps mitigate a well-known problem of EEND systems: when increasing the number of speakers the diarization performance drop is significantly reduced. We also introduce block-wise processing to be able to diarize arbitrarily long meetings., Comment: ICASSP 2023

Real-world visual data exhibit intrinsic hierarchical structures that can be represented effectively in hyperbolic spaces. Hyperbolic neural networks (HNNs) are a promising approach for learning feature representations in such spaces. However, current HNNs in computer vision rely on Euclidean backbones and only project features to the hyperbolic space in the task heads, limiting their ability to fully leverage the benefits of hyperbolic geometry. To address this, we present HCNN, a fully hyperbolic convolutional neural network (CNN) designed for computer vision tasks. Based on the Lorentz model, we generalize fundamental components of CNNs and propose novel formulations of the convolutional layer, batch normalization, and multinomial logistic regression. {Experiments on standard vision tasks demonstrate the promising performance of our HCNN framework in both hybrid and fully hyperbolic settings.} Overall, we believe our contributions provide a foundation for developing more powerful HNNs that can better represent complex structures found in image data. Our code is publicly available at https://github.com/kschwethelm/HyperbolicCV.

The scope of speech enhancement has changed from a monolithic view of single, independent tasks, to a joint processing of complex conversational speech recordings. Training and evaluation of these single tasks requires synthetic data with access to intermediate signals that is as close as possible to the evaluation scenario. As such data often is not available, many works instead use specialized databases for the training of each system component, e.g WSJ0-mix for source separation. We present a Multi-purpose Multi-Speaker Mixture Signal Generator (MMS-MSG) for generating a variety of speech mixture signals based on any speech corpus, ranging from classical anechoic mixtures (e.g., WSJ0-mix) over reverberant mixtures (e.g., SMS-WSJ) to meeting-style data. Its highly modular and flexible structure allows for the simulation of diverse environments and dynamic mixing, while simultaneously enabling an easy extension and modification to generate new scenarios and mixture types. These meetings can be used for prototyping, evaluation, or training purposes. We provide example evaluation data and baseline results for meetings based on the WSJ corpus. Further, we demonstrate the usefulness for realistic scenarios by using MMS-MSG to provide training data for the LibriCSS database., Comment: Accepted at IWAENC 2022

We propose a system that transcribes the conversation of a typical meeting scenario that is captured by a set of initially unsynchronized microphone arrays at unknown positions. It consists of subsystems for signal synchronization, including both sampling rate and sampling time offset estimation, diarization based on speaker and microphone array position estimation, multi-channel speech enhancement, and automatic speech recognition. With the estimated diarization information, a spatial mixture model is initialized that is used to estimate beamformer coefficients for source separation. Simulations show that the speech recognition accuracy can be improved by synchronizing and combining multiple distributed microphone arrays compared to a single compact microphone array. Furthermore, the proposed informed initialization of the spatial mixture model delivers a clear performance advantage over random initialization., Comment: Submitted to INTERSPEECH 2022

Spatial mixture model (SMM) supported acoustic beamforming has been extensively used for the separation of simultaneously active speakers. However, it has hardly been considered for the separation of meeting data, that are characterized by long recordings and only partially overlapping speech. In this contribution, we show that the fact that often only a single speaker is active can be utilized for a clever initialization of an SMM that employs time-varying class priors. In experiments on LibriCSS we show that the proposed initialization scheme achieves a significantly lower Word Error Rate (WER) on a downstream speech recognition task than a random initialization of the class probabilities by drawing from a Dirichlet distribution. With the only requirement that the number of speakers has to be known, we obtain a WER of 5.9 %, which is comparable to the best reported WER on this data set. Furthermore, the estimated speaker activity from the mixture model serves as a diarization based on spatial information., Comment: Submitted to INTERSPEECH 2022

We present a Reduced Order Model (ROM) which exploits recent developments in Physics Informed Neural Networks (PINNs) for solving inverse problems for the Navier--Stokes equations (NSE). In the proposed approach, the presence of simulated data for the fluid dynamics fields is assumed. A POD-Galerkin ROM is then constructed by applying POD on the snapshots matrices of the fluid fields and performing a Galerkin projection of the NSE (or the modified equations in case of turbulence modeling) onto the POD reduced basis. A $\textit{POD-Galerkin PINN ROM}$ is then derived by introducing deep neural networks which approximate the reduced outputs with the input being time and/or parameters of the model. The neural networks incorporate the physical equations (the POD-Galerkin reduced equations) into their structure as part of the loss function. Using this approach, the reduced model is able to approximate unknown parameters such as physical constants or the boundary conditions. A demonstration of the applicability of the proposed ROM is illustrated by two cases which are the steady flow around a backward step and the unsteady turbulent flow around a surface mounted cubic obstacle., Comment: Second version with additional test case

Impressive progress in neural network-based single-channel speech source separation has been made in recent years. But those improvements have been mostly reported on anechoic data, a situation that is hardly met in practice. Taking the SepFormer as a starting point, which achieves state-of-the-art performance on anechoic mixtures, we gradually modify it to optimize its performance on reverberant mixtures. Although this leads to a word error rate improvement by 7 percentage points compared to the standard SepFormer implementation, the system ends up with only marginally better performance than a PIT-BLSTM separation system, that is optimized with rather straightforward means. This is surprising and at the same time sobering, challenging the practical usefulness of many improvements reported in recent years for monaural source separation on nonreverberant data., Comment: Submitted to IWAENC 2022

Using machine learning (ML) techniques in general and deep learning techniques in specific needs a certain amount of data often not available in large quantities in technical domains. The manual inspection of machine tool components and the manual end-of-line check of products are labor-intensive tasks in industrial applications that companies often want to automate. To automate classification processes and develop reliable and robust machine learning-based classification and wear prognostics models, one needs real-world datasets to train and test the models. The dataset is available under https://doi.org/10.5445/IR/1000129520., Comment: 7 pages, 13 figures

The rapid uptake of renewable energy technologies in recent decades has increased the demand of energy researchers, policymakers and energy planners for reliable data on the spatial distribution of their costs and potentials. For onshore wind energy this has resulted in an active research field devoted to analysing these resources for regions, countries or globally. A particular thread of this research attempts to go beyond purely technical or spatial restrictions and determine the realistic, feasible or actual potential for wind energy. Motivated by these developments, this paper reviews methods and assumptions for analysing geographical, technical, economic and, finally, feasible onshore wind potentials. We address each of these potentials in turn, including aspects related to land eligibility criteria, energy meteorology, and technical developments relating to wind turbine characteristics such as power density, specific rotor power and spacing aspects. Economic aspects of potential assessments are central to future deployment and are discussed on a turbine and system level covering levelized costs depending on locations, and the system integration costs which are often overlooked in such analyses. Non-technical approaches include scenicness assessments of the landscape, expert and stakeholder workshops, willingness to pay / accept elicitations and socioeconomic cost-benefit studies. For each of these different potential estimations, the state of the art is critically discussed, with an attempt to derive best practice recommendations and highlight avenues for future research.

In this work we address disentanglement of style and content in speech signals. We propose a fully convolutional variational autoencoder employing two encoders: a content encoder and a style encoder. To foster disentanglement, we propose adversarial contrastive predictive coding. This new disentanglement method does neither need parallel data nor any supervision. We show that the proposed technique is capable of separating speaker and content traits into the two different representations and show competitive speaker-content disentanglement performance compared to other unsupervised approaches. We further demonstrate an increased robustness of the content representation against a train-test mismatch compared to spectral features, when used for phone recognition., Comment: accepted by icassp 2021

Protecting embedded security is becoming an increasingly challenging research problem for embedded systems due to a number of emerging trends in hardware, software, networks, and applications. Without fundamental advances in, and an understanding of embedded security it will be difficult for future engineers to provide assurance for the Internet of Things (IoT) and Operational Technology (OT) in wide ranging applications, from home automation and autonomous transportation to medical devices and factory floors. Common to such applications are cyberphysical risks and consequences stemming from a lack of embedded security. The Computing Community Consortium (CCC) held a one-day visioning workshop to explore these issues. The workshop focused on five major application areas of embedded systems, namely (1) medical/wearable devices, (2) autonomous systems (drones, vehicles, robots), (3) smart homes, (4) industry and supply chain, and (5) critical infrastructure. This report synthesizes the results of that workshop and develops a list of strategic goals for research and education over the next 5-10 years. Embedded security in connected devices presents challenges that require a broad look at the overall systems design, including human and societal dimensions as well as technical. Particular issues related to embedded security are a subset of the overall security of the application areas, which must also balance other design criteria such as cost, power, reliability, usability and function. Recent trends are converging to make the security of embedded systems an increasingly important and difficult objective, requiring new trans-disciplinary approaches to solve problems on a 5-10 year horizon., Comment: A Computing Community Consortium (CCC) workshop report, 24 pages

Deep metric learning employs deep neural networks to embed instances into a metric space such that distances between instances of the same class are small and distances between instances from different classes are large. In most existing deep metric learning techniques, the embedding of an instance is given by a feature vector produced by a deep neural network and Euclidean distance or cosine similarity defines distances between these vectors. In this paper, we study deep distributional embeddings of sequences, where the embedding of a sequence is given by the distribution of learned deep features across the sequence. This has the advantage of capturing statistical information about the distribution of patterns within the sequence in the embedding. When embeddings are distributions rather than vectors, measuring distances between embeddings involves comparing their respective distributions. We propose a distance metric based on Wasserstein distances between the distributions and a corresponding loss function for metric learning, which leads to a novel end-to-end trainable embedding model. We empirically observe that distributional embeddings outperform standard vector embeddings and that training with the proposed Wasserstein metric outperforms training with other distance functions.

We study the problem of inferring readers' identities and estimating their level of text comprehension from observations of their eye movements during reading. We develop a generative model of individual gaze patterns (scanpaths) that makes use of lexical features of the fixated words. Using this generative model, we derive a Fisher-score representation of eye-movement sequences. We study whether a Fisher-SVM with this Fisher kernel and several reference methods are able to identify readers and estimate their level of text comprehension based on eye-tracking data. While none of the methods are able to estimate text comprehension accurately, we find that the SVM with Fisher kernel excels at identifying readers., Comment: Proceedings of the European Conference on Machine Learning, 2018