This research investigates reliable local explanations for machine listening models.
problem Generating reliable local explanations for machine listening models.
method Investigates the sensitivity of SoundLIME explanations to input perturbations and proposes a novel method for identifying suitable content types.
result SoundLIME explanations are sensitive to the content in occluded input regions, and the average magnitude of input mel-spectrogram bins is the most suitable content type for temporal explanations.
Optimizes explanations for better listener understanding.
problem Insufficient consideration of listener preferences in concept-based explanations.
method Iterative training procedure based on direct preference optimization.
result Pragmatic explanations improve both model accuracy and user understanding.
CNNs improve generalization to unseen audio devices with increased width, not depth.
problem CNNs are sensitive to specific audio recording devices in acoustic scene classification.
method Investigated the relationship between over-parameterization and generalization in CNNs for audio classification.
result Increasing width improves generalization to unseen devices without increasing the number of parameters.
Industrial-scale podcast recommender system optimizes long-term listening journeys.
problem Optimizing long-term listening experiences in podcast recommendation systems.
method Reinforcement learning approach to optimize user listening journeys over months.
result Significantly improved long-term performance in A/B tests compared to short-term metrics.
OtoWorld helps agents learn to navigate by listening in interactive environments.
problem Training agents to navigate using auditory information.
method Interactive environment with agents learning to listen and navigate.
result Agents can win at OtoWorld, a navigation game with sound sources.
We present Listen, Attend and Spell (LAS), a neural network that learns to transcribe speech utterances to characters. Unlike traditional DNN-HMM models, this model learns all the components of a speech recognizer jointly. Our system has two components: a listener and a speller. The listener is a pyramidal recurrent ne…
Distribution mismatches between the data seen at training and at application time remain a major challenge in all application areas of machine learning. We study this problem in the context of machine listening (Task 1b of the DCASE 2019 Challenge). We propose a novel approach to learn domain-invariant classifiers in a…
This paper summarizes some recent advances on a set of tasks related to the processing of singing using state-of-the-art deep learning techniques. We discuss their achievements in terms of accuracy and sound quality, and the current challenges, such as availability of data and computing resources. We also discuss the i…
From a machine learning perspective, the human ability localize sounds can be modeled as a non-parametric and non-linear regression problem between binaural spectral features of sound received at the ears (input) and their sound-source directions (output). The input features can be summarized in terms of the individual…
We examine the effects of instantiating Lewis signaling games within a population of speaker and listener agents with the aim of producing a set of general and robust representations of unstructured pixel data. Preliminary experiments suggest that the set of representations associated with languages generated within a …
South Indian classical music (Carnatic music) is best consumed through live concerts. A carnatic recital requires meticulous planning accounting for several parameters like the performers' repertoire, composition variety, musical versatility, thematic structure, the recital's arrangement, etc. to ensure that the audien…
Trains word embeddings from music and text data to link music contexts.
problem Varying vocabulary size and musical relevance in word embeddings.
method Combines general text and music-specific data to train word embeddings.
result Trained embeddings better associate music contexts with compositions.
Podcast recommendations improved by analyzing user listening paths.
problem Challenges in recommending podcasts effectively.
method Analyzes user listening paths as sequential trajectories for recommendations.
result 450% increase in effectiveness over baseline.
In automatic speech recognition (ASR) what a user says depends on the particular context she is in. Typically, this context is represented as a set of word n-grams. In this work, we present a novel, all-neural, end-to-end (E2E) ASR sys- tem that utilizes such context. Our approach, which we re- fer to as Contextual Lis…
In a noisy environment, a lossy speech signal can be automatically restored by a listener if he/she knows the language well. That is, with the built-in knowledge of a "language model", a listener may effectively suppress noise interference and retrieve the target speech signals. Accordingly, we argue that familiarity w…
Emotion recognition based on EEG has become an active research area. As one of the machine learning models, CNN has been utilized to solve diverse problems including issues in this domain. In this work, a study of CNN and its spatiotemporal feature extraction has been conducted in order to explore capabilities of the m…
Deep-neural-network (DNN) based noise suppression systems yield significant improvements over conventional approaches such as spectral subtraction and non-negative matrix factorization, but do not generalize well to noise conditions they were not trained for. In comparison to DNNs, humans show remarkable noise suppress…
Heart diseases constitute a global health burden, and the problem is exacerbated by the error-prone nature of listening to and interpreting heart sounds. This motivates the development of automated classification to screen for abnormal heart sounds. Existing machine learning-based systems achieve accurate classificatio…
To a compact hyperbolic Riemann surface, we associate a finitely summable spectral triple whose underlying topological space is the limit set of a corresponding Schottky group, and whose ``Riemannian'' aspect (Hilbert space and Dirac operator) encode the boundary action through its Patterson-Sullivan measure. We prove …
Speech synthesis from EEG features using RNN.
problem Speech synthesis from EEG data.
method Recurrent Neural Network (RNN) regression model to predict acoustic features from EEG features.
result Feasibility of synthesizing speech directly from EEG features demonstrated.
Generates music with coherent rhythm, chords, and melody using LSTM models.
problem Lack of direction and coherence in generated music by neural networks.
method Two-stage LSTM model: first generates harmonic and rhythmic templates, then melodies conditioned on these.
result Subjective test shows improved musical coherence and coherence compared to baselines.
We study association between macroeconomic news and stock market returns using the statistical theory of copulas, and a new comprehensive measure of news based on the indexing of news wires. We find the impact of economic news on equity returns to be nonlinear and asymmetric. In particular, controlling for economic con…
We uniquely and explicitly reconstruct the instantaneous intrinsic metric of the Kerr-Newman Event Horizon from the spectrum of its Laplacian. In the process we find that the angular momentum parameter, radius, area; and in the uncharged case, mass, can be written in terms of these eigenvalues. In the uncharged case th…
Articulatory distinctive features, as well as phonetic transcription, play important role in speech-related tasks: computer-assisted pronunciation training, text-to-speech conversion (TTS), studying speech production mechanisms, speech recognition for low-resourced languages. End-to-end approaches to speech-related tas…
We address the problem of combining sequence models of symbolic music with user defined constraints. For typical models this is non-trivial as only the conditional distribution of each symbol given the earlier symbols is available, while the constraints correspond to arbitrary times. Previously this has been addressed …
With millimeter wave wireless communications, the resulting radiation reflects on most visible objects, creating rich multipath environments, namely in urban scenarios. The radiation captured by a listening device is thus shaped by the obstacles encountered, which carry latent information regarding their relative posit…
We propose a method to generate audio adversarial examples that can attack a state-of-the-art speech recognition model in the physical world. Previous work assumes that generated adversarial examples are directly fed to the recognition model, and is not able to perform such a physical attack because of reverberation an…
The musical notes from a hyperbolic marimba can identify the shape of hyperbolic surfaces.
problem Identifying hyperbolic surfaces based on their musical notes.
method Assigning musical notes to geodesics hitting labeled curves on hyperbolic surfaces.
result The melodies produced by hyperbolic marimbas can characterize hyperbolic surfaces up to isometry.
We consider the problem of learning high-level controls over the global structure of generated sequences, particularly in the context of symbolic music generation with complex language models. In this work, we present the Transformer autoencoder, which aggregates encodings of the input data across time to obtain a glob…
Classifier learns to ignore unreliable feedback from end users.
problem Improving classifier performance by filtering unreliable feedback.
method Modeling end users as autonomous agents, periodically retraining classifier with filtered feedback.
result Classifier can identify and filter out unreliable feedback, improving performance.
Which song will Smith listen to next? Which restaurant will Alice go to tomorrow? Which product will John click next? These applications have in common the prediction of user trajectories that are in a constant state of flux over a hidden network (e.g. website links, geographic location). What users are doing now may b…
The state-of-the-art in text-to-speech synthesis has recently improved considerably due to novel neural waveform generation methods, such as WaveNet. However, these methods suffer from their slow sequential inference process, while their parallel versions are difficult to train and even more expensive computationally. …
This paper will describe a novel approach to the cocktail party problem that relies on a fully convolutional neural network (FCN) architecture. The FCN takes noisy audio data as input and performs nonlinear, filtering operations to produce clean audio data of the target speech at the output. Our method learns a model f…
Generative Adversarial Networks (GANs) have become exceedingly popular in a wide range of data-driven research fields, due in part to their success in image generation. Their ability to generate new samples, often from only a small amount of input data, makes them an exciting research tool in areas with limited data re…
This paper introduces a model of environmental acoustic scenes which adopts a morphological approach by ab-stracting temporal structures of acoustic scenes. To demonstrate its potential, this model is employed to evaluate the performance of a large set of acoustic events detection systems. This model allows us to expli…
This paper proposes a framework which is able to generate a sequence of three-dimensional human dance poses for a given music. The proposed framework consists of three components: a music feature encoder, a pose generator, and a music genre classifier. We focus on integrating these components for generating a realistic…
Having a sequence-to-sequence model which can operate in an online fashion is important for streaming applications such as Voice Search. Neural transducer is a streaming sequence-to-sequence model, but has shown a significant degradation in performance compared to non-streaming models such as Listen, Attend and Spell (…
Sequence-to-sequence text-to-speech (TTS) is dominated by soft-attention-based methods. Recently, hard-attention-based methods have been proposed to prevent fatal alignment errors, but their sampling method of discrete alignment is poorly investigated. This research investigates various combinations of sampling methods…
We present a deep neural network-based methodology for synthesising percussive sounds with control over high-level timbral characteristics of the sounds. This approach allows for intuitive control of a synthesizer, enabling the user to shape sounds without extensive knowledge of signal processing. We use a feedforward …
In this paper, we investigate the common scenario where every candidate item for recommendation is characterized by a maximum capacity, i.e., number of seats in a Point-of-Interest (POI) or size of an item's inventory. Despite the prevalence of the task of recommending items under capacity constraints in a variety of s…
AV-ASR system improves speech recognition with visual context.
problem Improving speech recognition accuracy with visual information.
method Transformer-based architecture with multiresolution and multimodal training.
result Multiresolution training speeds up convergence and improves WER by 18%.
Nowadays, CAPTCHAs are computer generated tests that human can pass but current computer systems can not. They have common usage in various web services in order to be able to detect a human from computer programs autonomously. In this way, owners can protect their web services from bots. In addition to visual CAPTCHAs…
Deep neural networks (DNN) are able to successfully process and classify speech utterances. However, understanding the reason behind a classification by DNN is difficult. One such debugging method used with image classification DNNs is activation maximization, which generates example-images that are classified as one o…
We investigated the impact of noisy linguistic features on the performance of a Japanese speech synthesis system based on neural network that uses WaveNet vocoder. We compared an ideal system that uses manually corrected linguistic features including phoneme and prosodic information in training and test sets against a …
Brain networks in fMRI are typically identified using spatial independent component analysis (ICA), yet mathematical constraints such as sparse coding and positivity both provide alternate biologically-plausible frameworks for generating brain networks. Non-negative Matrix Factorization (NMF) would suppress negative BO…
Recent speech technology research has seen a growing interest in using WaveNets as statistical vocoders, i.e., generating speech waveforms from acoustic features. These models have been shown to improve the generated speech quality over classical vocoders in many tasks, such as text-to-speech synthesis and voice conver…
One of the challenges in computational acoustics is the identification of models that can simulate and predict the physical behavior of a system generating an acoustic signal. Whenever such models are used for commercial applications an additional constraint is the time-to-market, making automation of the sound design …
Paper uses SSL models' uncertainty to predict audio quality efficiently.
problem Efficiently predicting audio quality in low-resource settings.
method Leverages self-supervised learning models' uncertainty measures.
result Uncertainty measures correlate with MOS scores in SSL models.