Search

The performance of deep neural network-based speech enhancement systems typically increases with the training dataset size. However, studies that investigated the effect of training dataset size on speech enhancement performance did not consider recent approaches, such as diffusion-based generative models. Diffusion models are typically trained with massive datasets for image generation tasks, but whether this is also required for speech enhancement is unknown. Moreover, studies that investigated the effect of training dataset size did not control for the data diversity. It is thus unclear whether the performance improvement was due to the increased dataset size or diversity. Therefore, we systematically investigate the effect of training dataset size on the performance of popular state-of-the-art discriminative and diffusion-based speech enhancement systems. We control for the data diversity by using a fixed set of speech utterances, noise segments and binaural room impulse responses to generate datasets of different sizes. We find that the diffusion-based systems do not benefit from increasing the training dataset size as much as the discriminative systems. They perform the best relative to the discriminative systems with datasets of 10 h or less, but they are outperformed by the discriminative systems with datasets of 100 h or more.

Ensuring intelligible speech communication for hearing assistive devices in low-latency scenarios presents significant challenges in terms of speech enhancement, coding and transmission. In this paper, we propose novel solutions for low-latency joint speech transmission and enhancement, leveraging deep neural networks (DNNs). Our approach integrates two state-of-the-art DNN architectures for low-latency speech enhancement and low-latency analog joint source-channel-based transmission, creating a combined low-latency system and jointly training both systems in an end-to-end approach. Due to the computational demands of the enhancement system, this order is suitable when high computational power is unavailable in the decoder, like hearing assistive devices. The proposed system enables the configuration of total latency, achieving high performance even at latencies as low as 3 ms, which is typically challenging to attain. The simulation results provide compelling evidence that a joint enhancement and transmission system is superior to a simple concatenation system in diverse settings, encompassing various wireless channel conditions, latencies, and background noise scenarios.

Advanced auditory models are useful in designing signal-processing algorithms for hearing-loss compensation or speech enhancement. Such auditory models provide rich and detailed descriptions of the auditory pathway, and might allow for individualization of signal-processing strategies, based on physiological measurements. However, these auditory models are often computationally demanding, requiring significant time to compute. To address this issue, previous studies have explored the use of deep neural networks to emulate auditory models and reduce inference time. While these deep neural networks offer impressive efficiency gains in terms of computational time, they may suffer from uneven emulation performance as a function of auditory-model frequency-channels and input sound pressure level, making them unsuitable for many tasks. In this study, we demonstrate that the conventional machine-learning optimization objective used in existing state-of-the-art methods is the primary source of this limitation. Specifically, the optimization objective fails to account for the frequency- and level-dependencies of the auditory model, caused by a large input dynamic range and different types of hearing losses emulated by the auditory model. To overcome this limitation, we propose a new optimization objective that explicitly embeds the frequency- and level-dependencies of the auditory model. Our results show that this new optimization objective significantly improves the emulation performance of deep neural networks across relevant input sound levels and auditory-model frequency channels, without increasing the computational load during inference. Addressing these limitations is essential for advancing the application of auditory models in signal-processing tasks, ensuring their efficacy in diverse scenarios., Comment: Accepted by IEEE/ACM Transactions on Audio, Speech and Language Processing. This version is the authors' version and may vary from the final publication in details

This article investigates the use of deep neural networks (DNNs) for hearing-loss compensation. Hearing loss is a prevalent issue affecting millions of people worldwide, and conventional hearing aids have limitations in providing satisfactory compensation. DNNs have shown remarkable performance in various auditory tasks, including speech recognition, speaker identification, and music classification. In this study, we propose a DNN-based approach for hearing-loss compensation, which is trained on the outputs of hearing-impaired and normal-hearing DNN-based auditory models in response to speech signals. First, we introduce a framework for emulating auditory models using DNNs, focusing on an auditory-nerve model in the auditory pathway. We propose a linearization of the DNN-based approach, which we use to analyze the DNN-based hearing-loss compensation. Additionally we develop a simple approach to choose the acoustic center frequencies of the auditory model used for the compensation strategy. Finally, we evaluate the DNN-based hearing-loss compensation strategies using listening tests with hearing impaired listeners. The results demonstrate that the proposed approach results in feasible hearing-loss compensation strategies. Our proposed approach was shown to provide an increase in speech intelligibility and was found to outperform a conventional approach in terms of perceived speech quality.

Studies have shown that in noisy acoustic environments, providing binaural signals to the user of an assistive listening device may improve speech intelligibility and spatial awareness. This paper presents a binaural speech enhancement method using a complex convolutional neural network with an encoder-decoder architecture and a complex multi-head attention transformer. The model is trained to estimate individual complex ratio masks in the time-frequency domain for the left and right-ear channels of binaural hearing devices. The model is trained using a novel loss function that incorporates the preservation of spatial information along with speech intelligibility improvement and noise reduction. Simulation results for acoustic scenarios with a single target speaker and isotropic noise of various types show that the proposed method improves the estimated binaural speech intelligibility and preserves the binaural cues better in comparison with several baseline algorithms., Comment: Accepted to ICASSP 2024

In this paper, we propose the use of self-supervised pretraining on a large unlabelled data set to improve the performance of a personalized voice activity detection (VAD) model in adverse conditions. We pretrain a long short-term memory (LSTM)-encoder using the autoregressive predictive coding (APC) framework and fine-tune it for personalized VAD. We also propose a denoising variant of APC, with the goal of improving the robustness of personalized VAD. The trained models are systematically evaluated on both clean speech and speech contaminated by various types of noise at different SNR-levels and compared to a purely supervised model. Our experiments show that self-supervised pretraining not only improves performance in clean conditions, but also yields models which are more robust to adverse conditions compared to purely supervised learning., Comment: To be published at ICASSP2024, 14th of April 2024, Seoul, South Korea. Copyright (c) 2023 IEEE. 5 pages, 2, figures, 5 tables

Diffusion models are a new class of generative models that have shown outstanding performance in image generation literature. As a consequence, studies have attempted to apply diffusion models to other tasks, such as speech enhancement. A popular approach in adapting diffusion models to speech enhancement consists in modelling a progressive transformation between the clean and noisy speech signals. However, one popular diffusion model framework previously laid in image generation literature did not account for such a transformation towards the system input, which prevents from relating the existing diffusion-based speech enhancement systems with the aforementioned diffusion model framework. To address this, we extend this framework to account for the progressive transformation between the clean and noisy speech signals. This allows us to apply recent developments from image generation literature, and to systematically investigate design aspects of diffusion models that remain largely unexplored for speech enhancement, such as the neural network preconditioning, the training loss weighting, the stochastic differential equation (SDE), or the amount of stochasticity injected in the reverse process. We show that the performance of previous diffusion-based speech enhancement systems cannot be attributed to the progressive transformation between the clean and noisy speech signals. Moreover, we show that a proper choice of preconditioning, training loss weighting, SDE and sampler allows to outperform a popular diffusion-based speech enhancement system in terms of perceptual metrics while using fewer sampling steps, thus reducing the computational cost by a factor of four.

Diffusion models are a new class of generative models that have recently been applied to speech enhancement successfully. Previous works have demonstrated their superior performance in mismatched conditions compared to state-of-the art discriminative models. However, this was investigated with a single database for training and another one for testing, which makes the results highly dependent on the particular databases. Moreover, recent developments from the image generation literature remain largely unexplored for speech enhancement. These include several design aspects of diffusion models, such as the noise schedule or the reverse sampler. In this work, we systematically assess the generalization performance of a diffusion-based speech enhancement model by using multiple speech, noise and binaural room impulse response (BRIR) databases to simulate mismatched acoustic conditions. We also experiment with a noise schedule and a sampler that have not been applied to speech enhancement before. We show that the proposed system substantially benefits from using multiple databases for training, and achieves superior performance compared to state-of-the-art discriminative models in both matched and mismatched conditions. We also show that a Heun-based sampler achieves superior performance at a smaller computational cost compared to a sampler commonly used for speech enhancement., Comment: Accepted to ICASSP 2024

We consider speech enhancement for signals picked up in one noisy environment that must be rendered to a listener in another noisy environment. For both far-end noise reduction and near-end listening enhancement, it has been shown that excessive focus on noise suppression or intelligibility maximization may lead to excessive speech distortions and quality degradations in favorable noise conditions, where intelligibility is already at ceiling level. Recently [1,2] propose to remedy this with a minimum processing framework that either reduces noise or enhances listening a minimum amount given that a certain intelligibility criterion is still satisfied Additionally, it has been shown that joint consideration of both environments improves speech enhancement performance. In this paper, we formulate a joint far- and near-end minimum processing framework, that improves intelligibility while limiting speech distortions in favorable noise conditions. We provide closed-form solutions to specific boundary scenarios and investigate performance for the general case using numerical optimization. We also show concatenating existing minimum processing far- and near-end enhancement methods preserves the effects of the initial methods. Results show that the joint optimization can further improve performance compared to the concatenated approach., Comment: Accepted at IEEE ICASSP 2024 Workshop on Hands-free Speech Communication and Microphone Arrays (HSCMA) 2024

Audio and visual modalities are inherently connected in speech signals: lip movements and facial expressions are correlated with speech sounds. This motivates studies that incorporate the visual modality to enhance an acoustic speech signal or even restore missing audio information. Specifically, this paper focuses on the problem of audio-visual speech inpainting, which is the task of synthesizing the speech in a corrupted audio segment in a way that it is consistent with the corresponding visual content and the uncorrupted audio context. We present an audio-visual transformer-based deep learning model that leverages visual cues that provide information about the content of the corrupted audio. It outperforms the previous state-of-the-art audio-visual model and audio-only baselines. We also show how visual features extracted with AV-HuBERT, a large audio-visual transformer for speech recognition, are suitable for synthesizing speech., Comment: Accepted in Interspeech23

Distributed signal-processing algorithms in (wireless) sensor networks often aim to decentralize processing tasks to reduce communication cost and computational complexity or avoid reliance on a single device (i.e., fusion center) for processing. In this contribution, we extend a distributed adaptive algorithm for blind system identification that relies on the estimation of a stacked network-wide consensus vector at each node, the computation of which requires either broadcasting or relaying of node-specific values (i.e., local vector norms) to all other nodes. The extended algorithm employs a distributed-averaging-based scheme to estimate the network-wide consensus norm value by only using the local vector norm provided by neighboring sensor nodes. We introduce an adaptive mixing factor between instantaneous and recursive estimates of these norms for adaptivity in a time-varying system. Simulation results show that the extension provides estimation results close to the optimal fully-connected-network or broadcasting case while reducing inter-node transmission significantly., Comment: Accepted to ICASSP 2023

In this work, we propose a frequency bin-wise method to estimate the single-channel speech presence probability (SPP) with multiple deep neural networks (DNNs) in the short-time Fourier transform domain. Since all frequency bins are typically considered simultaneously as input features for conventional DNN-based SPP estimators, high model complexity is inevitable. To reduce the model complexity and the requirements on the training data, we take a single frequency bin and some of its neighboring frequency bins into account to train separate gate recurrent units. In addition, the noisy speech and the a posteriori probability SPP representation are used to train our model. The experiments were performed on the Deep Noise Suppression challenge dataset. The experimental results show that the speech detection accuracy can be improved when we employ the frequency bin-wise model. Finally, we also demonstrate that our proposed method outperforms most of the state-of-the-art SPP estimation methods in terms of speech detection accuracy and model complexity., Comment: Accepted for ICASSP 2023

In the context of keyword spotting (KWS), the replacement of handcrafted speech features by learnable features has not yielded superior KWS performance. In this study, we demonstrate that filterbank learning outperforms handcrafted speech features for KWS whenever the number of filterbank channels is severely decreased. Reducing the number of channels might yield certain KWS performance drop, but also a substantial energy consumption reduction, which is key when deploying common always-on KWS on low-resource devices. Experimental results on a noisy version of the Google Speech Commands Dataset show that filterbank learning adapts to noise characteristics to provide a higher degree of robustness to noise, especially when dropout is integrated. Thus, switching from typically used 40-channel log-Mel features to 8-channel learned features leads to a relative KWS accuracy loss of only 3.5% while simultaneously achieving a 6.3x energy consumption reduction.

The intelligibility and quality of speech from a mobile phone or public announcement system are often affected by background noise in the listening environment. By pre-processing the speech signal it is possible to improve the speech intelligibility and quality -- this is known as near-end listening enhancement (NLE). Although, existing NLE techniques are able to greatly increase intelligibility in harsh noise environments, in favorable noise conditions the intelligibility of speech reaches a ceiling where it cannot be further enhanced. Actually, the focus of existing methods solely on improving the intelligibility causes unnecessary processing of the speech signal and leads to speech distortions and quality degradations. In this paper, we provide a new rationale for NLE, where the target speech is minimally processed in terms of a processing penalty, provided that a certain performance constraint, e.g., intelligibility, is satisfied. We present a closed-form solution for the case where the performance criterion is an intelligibility estimator based on the approximated speech intelligibility index and the processing penalty is the mean-square error between the processed and the clean speech. This produces an NLE method that adapts to changing noise conditions via a simple gain rule by limiting the processing to the minimum necessary to achieve a desired intelligibility, while at the same time focusing on quality in favorable noise situations by minimizing the amount of speech distortions. Through simulation studies, we show the proposed method attains speech quality on par or better than existing methods in both objective measurements and subjective listening tests, whilst still sustaining objective speech intelligibility performance on par with existing methods.

We apply a Gaussian state formalism to track fluctuating perturbations that act on the position and momentum quadrature variables of a harmonic oscillator. Following a seminal proposal by Tsang and Caves [Phys. Rev. Lett. 105, 123601 (2010)], Einstein-Podolsky-Rosen correlations with the quadrature variables of an ancillary harmonic oscillator are leveraged to significantly improve the estimates as relevant sensor variables can be arbitrarily squeezed while evading adverse effects from the conjugate, anti-squeezed variables. Our real-time analysis of the continuous monitoring of the system employs a hybrid quantum-classical description of the quantum probe and the unknown classical perturbations, and it provides a general formalism to establish the achievements of the sensing scheme and how they depend on different parameters.

Spoken keyword spotting (KWS) deals with the identification of keywords in audio streams and has become a fast-growing technology thanks to the paradigm shift introduced by deep learning a few years ago. This has allowed the rapid embedding of deep KWS in a myriad of small electronic devices with different purposes like the activation of voice assistants. Prospects suggest a sustained growth in terms of social use of this technology. Thus, it is not surprising that deep KWS has become a hot research topic among speech scientists, who constantly look for KWS performance improvement and computational complexity reduction. This context motivates this paper, in which we conduct a literature review into deep spoken KWS to assist practitioners and researchers who are interested in this technology. Specifically, this overview has a comprehensive nature by covering a thorough analysis of deep KWS systems (which includes speech features, acoustic modeling and posterior handling), robustness methods, applications, datasets, evaluation metrics, performance of deep KWS systems and audio-visual KWS. The analysis performed in this paper allows us to identify a number of directions for future research, including directions adopted from automatic speech recognition research and directions that are unique to the problem of spoken KWS.

In this paper, we propose a method to estimate the proximity of an acoustic reflector, e.g., a wall, using ego-noise, i.e., the noise produced by the moving parts of a listening robot. This is achieved by estimating the times of arrival of acoustic echoes reflected from the surface. Simulated experiments show that the proposed nonintrusive approach is capable of accurately estimating the distance of a reflector up to 1 meter and outperforms a previously proposed intrusive approach under loud ego-noise conditions. The proposed method is helped by a probabilistic echo detector that estimates whether or not an acoustic reflector is within a short range of the robotic platform. This preliminary investigation paves the way towards a new kind of collision avoidance system that would purely rely on audio sensors rather than conventional proximity sensors.

This paper considers speech enhancement of signals picked up in one noisy environment which must be presented to a listener in another noisy environment. Recently, it has been shown that an optimal solution to this problem requires the consideration of the noise sources in both environments jointly. However, the existing optimal mutual information based method requires a complicated system model that includes natural speech variations, and relies on approximations and assumptions of the underlying signal distributions. In this paper, we propose to use a simpler signal model and optimize speech intelligibility based on the Approximated Speech Intelligibility Index (ASII). We derive a closed-form solution to the joint far- and near-end speech enhancement problem that is independent of the marginal distribution of signal coefficients, and that achieves similar performance to existing work. In addition, we do not need to model or optimize for natural speech variations.

Knowledge of quantum mechanical systems is becoming more important for many science and engineering students who are looking to join the emerging quantum workforce. To better prepare a wide range of students for these careers, we must seek to develop new tools to enhance our education in quantum topics. We present initial studies on the use of one of these such tools, Quantum Composer, a 1D quantum simulation and visualization tool developed for education and research purposes. In particular, we conducted five think-aloud interviews with students who worked through an exercise using Quantum Composer that focused on the statics and dynamics of quantum states in a single harmonic well system. Our results show that Quantum Composer helps students to obtain the correct answers to the questions posed, but additional support is needed to facilitate the development of student reasoning behind these answers. We also show that students are able to focus only on the relevant features of Quantum Composer to achieve the task.

In recent years, speech processing algorithms have seen tremendous progress primarily due to the deep learning renaissance. This is especially true for speech separation where the time-domain audio separation network (TasNet) has led to significant improvements. However, for the related task of single-speaker speech enhancement, which is of obvious importance, it is yet unknown, if the TasNet architecture is equally successful. In this paper, we show that TasNet improves state-of-the-art also for speech enhancement, and that the largest gains are achieved for modulated noise sources such as speech. Furthermore, we show that TasNet learns an efficient inner-domain representation, where target and noise signal components are highly separable. This is especially true for noise in terms of interfering speech signals, which might explain why TasNet performs so well on the separation task. Additionally, we show that TasNet performs poorly for large frame hops and conjecture that aliasing might be the main cause of this performance drop. Finally, we show that TasNet consistently outperforms a state-of-the-art single-speaker speech enhancement system.

Development of outreach skills is critical for researchers when communicating their work to non-expert audiences. However, due to the lack of formal training, researchers are typically unaware of the benefits of outreach training and often under-prioritize outreach. We present a training programme conducted with an international network of PhD students in quantum physics, which focused on developing outreach skills and an understanding of the associated professional benefits by creating an outreach portfolio consisting of a range of implementable outreach products. We describe our approach, assess the impact, and provide a list of guidelines for designing similar programmes across scientific disciplines in the future., Comment: 14 figures, 3 tables

In this paper, we present a deep-learning-based framework for audio-visual speech inpainting, i.e., the task of restoring the missing parts of an acoustic speech signal from reliable audio context and uncorrupted visual information. Recent work focuses solely on audio-only methods and generally aims at inpainting music signals, which show highly different structure than speech. Instead, we inpaint speech signals with gaps ranging from 100 ms to 1600 ms to investigate the contribution that vision can provide for gaps of different duration. We also experiment with a multi-task learning approach where a phone recognition task is learned together with speech inpainting. Results show that the performance of audio-only speech inpainting approaches degrades rapidly when gaps get large, while the proposed audio-visual approach is able to plausibly restore missing information. In addition, we show that multi-task learning is effective, although the largest contribution to performance comes from vision., Comment: Accepted at ICASSP 2021

Speech enhancement and speech separation are two related tasks, whose purpose is to extract either one or more target speech signals, respectively, from a mixture of sounds generated by several sources. Traditionally, these tasks have been tackled using signal processing and machine learning techniques applied to the available acoustic signals. Since the visual aspect of speech is essentially unaffected by the acoustic environment, visual information from the target speakers, such as lip movements and facial expressions, has also been used for speech enhancement and speech separation systems. In order to efficiently fuse acoustic and visual information, researchers have exploited the flexibility of data-driven approaches, specifically deep learning, achieving strong performance. The ceaseless proposal of a large number of techniques to extract features and fuse multimodal information has highlighted the need for an overview that comprehensively describes and discusses audio-visual speech enhancement and separation based on deep learning. In this paper, we provide a systematic survey of this research topic, focusing on the main elements that characterise the systems in the literature: acoustic features; visual features; deep learning methods; fusion techniques; training targets and objective functions. In addition, we review deep-learning-based methods for speech reconstruction from silent videos and audio-visual sound source separation for non-speech signals, since these methods can be more or less directly applied to audio-visual speech enhancement and separation. Finally, we survey commonly employed audio-visual speech datasets, given their central role in the development of data-driven approaches, and evaluation methods, because they are generally used to compare different systems and determine their performance.

We demonstrate the efficiency of a recent exact-gradient optimal control methodology by applying it to a challenging many-body problem, crossing the superfluid to Mott-insulator phase transition in the Bose-Hubbard model. The system size necessitates a matrix product state representation and this seamlessly integrates with the requirements of the algorithm. We observe fidelities in the range 0.99-0.9999 with associated minimal process duration estimates displaying an exponential fidelity-duration trade-off across several orders of magnitude. The corresponding optimal solutions are characterized in terms of a predominantly linear sweep across the critical point followed by bang-bang-like structure. This is quite different from the smooth and monotonic solutions identified by earlier gradient-free optimizations which are hampered in locating the higher complexity protocols in the regime of high-fidelities at low process durations. Overall, the comparison suggests significant methodological improvements also for many-body systems in the ideal open-loop setting. Acknowledging that idealized open-loop control may deteriorate in actual experiments, we discuss the merits of using such an approach in combination with closed-loop control -- in particular, high-fidelity physical insights extracted with the former can be used to formulate practical, low-dimensional search spaces for the latter., Comment: 19 pages, 9 figures. Content is partially a split revision of this V1 preprint arXiv:2005.09943v1

Making quantum mechanical equations and concepts come to life through interactive simulation and visualization are commonplace for augmenting learning and teaching. However, graphical visualizations nearly always exhibit a set of hard-coded functionalities while corresponding text-based codes offer a higher degree of flexibility at the expense of steep learning curves or time investments. We introduce Quantum Composer, which allows the user to build, expand, or explore quantum mechanical simulations by interacting with graphically connectable nodes, each corresponding to a physical concept, mathematical operation, visualization, etc. Abstracting away numerical and programming details while at the same time retaining accessibility, emphasis on understanding, and rapid feedback mechanisms, we illustrate through a series of examples its open-ended applicability in both introductory and advanced quantum mechanics courses, student projects, and for visual exploration within research environments., Comment: 29 pages, 12 figures

Efficient optimization of quantum systems is a necessity for reaching fault tolerant thresholds. A standard tool for optimizing simulated quantum dynamics is the gradient-based \textsc{grape} algorithm, which has been successfully applied in a wide range of different branches of quantum physics. In this work, we derive and implement exact $2^{\mathrm{nd}}$ order analytical derivatives of the coherent dynamics and find improvements compared to the standard of optimizing with the approximate $2^{\mathrm{nd}}$ order \textsc{bfgs}. We demonstrate performance improvements for both the best and average errors of constrained unitary gate synthesis on a circuit-\textsc{qed} system over a broad range of different gate durations., Comment: 13 pages, 5 figures

Despite their great performance over the years, handcrafted speech features are not necessarily optimal for any particular speech application. Consequently, with greater or lesser success, optimal filterbank learning has been studied for different speech processing tasks. In this paper, we fill in a gap by exploring filterbank learning for keyword spotting (KWS). Two approaches are examined: filterbank matrix learning in the power spectral domain and parameter learning of a psychoacoustically-motivated gammachirp filterbank. Filterbank parameters are optimized jointly with a modern deep residual neural network-based KWS back-end. Our experimental results reveal that, in general, there are no statistically significant differences, in terms of KWS accuracy, between using a learned filterbank and handcrafted speech features. Thus, while we conclude that the latter are still a wise choice when using modern KWS back-ends, we also hypothesize that this could be a symptom of information redundancy, which opens up new research possibilities in the field of small-footprint KWS.

Accurate derivatives are important for efficiently locally traversing and converging in quantum optimization landscapes. By deriving analytically exact control derivatives (gradient and Hessian) for unitary control tasks, we show here that the computational feasibility of meeting this accuracy requirement depends on the choice of propagation scheme and problem representation. Even when exact propagation is sufficiently cheap it is, perhaps surprisingly, much more efficient to optimize the (appropriately) approximate propagators: approximations in the dynamics are traded off for significant complexity reductions in the exact derivative calculations. Importantly, past the initial analytical considerations, only standard numerical techniques are explicitly required with straightforward application to realistic systems. These results are numerically verified for two concrete problems of increasing Hilbert space dimensionality. The best schemes obtain unit fidelity to machine precision whereas the results for other schemes are separated consistently by orders of magnitude in computation time and in worst case 10 orders of magnitude in achievable fidelity. Since these gaps continually increase with system size and complexity, this methodology allows numerically efficient optimization of very high-dimensional dynamics, e.g. in many-body contexts, operating in the high-fidelity regime which will be published separately., Comment: 20 pages, 4 figures. V2 and onwards has had content moved to another preprint here: arXiv:2008.06076

Citizen science methodologies have over the past decade been applied with great success to help solve highly complex numerical challenges. Here, we take early steps in the quantum physics arena by introducing a citizen science game, Quantum Moves 2, and compare the performance of different optimization methods across three different quantum optimal control problems of varying difficulty. Inside the game, players can apply a gradient-based algorithm (running locally on their device) to optimize their solutions and we find that these results perform roughly on par with the best of the tested standard optimization methods performed on a computer cluster. In addition, cluster-optimized player seeds was the only method to exhibit roughly optimal performance across all three challenges. This highlights the potential for crowdsourcing the solution of future quantum research problems., Comment: 15 figures

Both acoustic and visual information influence human perception of speech. For this reason, the lack of audio in a video sequence determines an extremely low speech intelligibility for untrained lip readers. In this paper, we present a way to synthesise speech from the silent video of a talker using deep learning. The system learns a mapping function from raw video frames to acoustic features and reconstructs the speech with a vocoder synthesis algorithm. To improve speech reconstruction performance, our model is also trained to predict text information in a multi-task learning fashion and it is able to simultaneously reconstruct and recognise speech in real time. The results in terms of estimated speech quality and intelligibility show the effectiveness of our method, which exhibits an improvement over existing video-to-speech approaches., Comment: Accepted to Interspeech 2020