Search Results for author:
Found 34 papers, 13 papers with code
Full-band General Audio Synthesis with Score-based Diffusion
Recent works have shown the capability of deep generative models to tackle general audio synthesis from a single label, producing a variety of impulsive, tonal, and environmental sounds.
Quantitative Evidence on Overlooked Aspects of Enrollment Speaker Embeddings for Target Speaker Separation
Single channel target speaker separation (TSS) aims at extracting a speaker's voice from a mixture of multiple talkers given an enrollment utterance of that speaker.
Adversarial Permutation Invariant Training for Universal Sound Separation
Universal sound separation consists of separating mixes with arbitrary sounds of different types, and permutation invariant training (PIT) is used to train source agnostic models that do so.
Universal Speech Enhancement with Score-based Diffusion
We hope that both our methodology and technical contributions encourage researchers and practitioners to adopt a universal approach to speech enhancement, possibly framing it as a generative task.
On loss functions and evaluation metrics for music source separation
We investigate which loss functions provide better separations via benchmarking an extensive set of those for music source separation.
Upsampling layers for music source separation
Upsampling artifacts are caused by problematic upsampling layers and due to spectral replicas that emerge while upsampling.
Assessing Algorithmic Biases for Musical Version Identification
Version identification (VI) systems now offer accurate and scalable solutions for detecting different renditions of a musical composition, allowing the use of these systems in industrial applications and throughout the wider music ecosystem.
On tuning consistent annealed sampling for denoising score matching
Score-based generative models provide state-of-the-art quality for image and audio synthesis.
Investigating the efficacy of music version retrieval systems for setlist identification
The setlist identification (SLI) task addresses a music recognition use case where the goal is to retrieve the metadata and timestamps for all the tracks played in live music events.
Upsampling artifacts in neural audio synthesis
We then compare different upsampling layers, showing that nearest neighbor upsamplers can be an alternative to the problematic (but state-of-the-art) transposed and subpixel convolutions which are prone to introduce tonal artifacts.
Automatic multitrack mixing with a differentiable mixing console of neural audio effects
Applications of deep learning to automatic multitrack mixing are largely unexplored.
Audio and Speech Processing Sound
Less is more: Faster and better music version identification with embedding distillation
Version identification systems aim to detect different renditions of the same underlying musical composition (loosely called cover songs).
SESQA: semi-supervised learning for speech quality assessment
Automatic speech quality assessment is an important, transversal task whose progress is hampered by the scarcity of human annotations, poor generalization to unseen recording conditions, and a lack of flexibility of existing approaches.
Accurate and Scalable Version Identification Using Musically-Motivated Embeddings
The version identification (VI) task deals with the automatic detection of recordings that correspond to the same underlying musical piece.
Ranked #4 on
Cover song identification
on YouTube350
Input complexity and out-of-distribution detection with likelihood-based generative models
Likelihood-based generative models are a promising resource to detect out-of-distribution (OOD) inputs which could compromise the robustness or reliability of a machine learning system.
Ranked #10 on
Anomaly Detection
on Unlabeled CIFAR-10 vs CIFAR-100
Blow: a single-scale hyperconditioned flow for non-parallel raw-audio voice conversion
End-to-end models for raw audio generation are a challenge, specially if they have to work with non-parallel data, which is a desirable setup in many situations.
Learning Problem-agnostic Speech Representations from Multiple Self-supervised Tasks
Learning good representations without supervision is still an open issue in machine learning, and is particularly challenging for speech signals, which are often characterized by long sequences with a complex hierarchical structure.
Ranked #2 on
Distant Speech Recognition
on DIRHA English WSJ
Towards Generalized Speech Enhancement with Generative Adversarial Networks
The speech enhancement task usually consists of removing additive noise or reverberation that partially mask spoken utterances, affecting their intelligibility.
Training neural audio classifiers with few data
We investigate supervised learning strategies that improve the training of neural network audio classifiers on small annotated collections.
Whispered-to-voiced Alaryngeal Speech Conversion with Generative Adversarial Networks
Most methods of voice restoration for patients suffering from aphonia either produce whispered or monotone speech.
Self-Attention Linguistic-Acoustic Decoder
The conversion from text to speech relies on the accurate mapping from linguistic to acoustic symbol sequences, for which current practice employs recurrent statistical models like recurrent neural networks.
Assessing the impact of machine intelligence on human behaviour: an interdisciplinary endeavour
The workshop gathered an interdisciplinary group of experts to establish the state of the art research in the field and a list of future research challenges to be addressed on the topic of human and machine intelligence, algorithm's potential impact on human cognitive capabilities and decision making, and evaluation and regulation needs.
Towards a universal neural network encoder for time series
We evaluate the performance of the proposed approach on a well-known time series classification benchmark, considering full adaptation, partial adaptation, and no adaptation of the encoder to the new data type.
Overcoming catastrophic forgetting with hard attention to the task
In this paper, we propose a task-based hard attention mechanism that preserves previous tasks' information without affecting the current task's learning.
Ranked #2 on
Continual Learning
on 20Newsgroup (10 tasks)
Continual Prediction of Notification Attendance with Classical and Deep Network Approaches
Besides using classical gradient-boosted trees, we demonstrate how to make continual predictions using a recurrent neural network (RNN).
Human-Computer Interaction
Language and Noise Transfer in Speech Enhancement Generative Adversarial Network
In this work, we present the results of adapting a speech enhancement generative adversarial network by finetuning the generator with small amounts of data.
Getting deep recommenders fit: Bloom embeddings for sparse binary input/output networks
Due to the structure of the data coming from recommendation domains (i. e., one-hot-encoded vectors of item preferences), these algorithms tend to have large input and output dimensionalities that dominate their overall size.
Practical Processing of Mobile Sensor Data for Continual Deep Learning Predictions
We present a practical approach for processing mobile sensor time series data for continual deep learning predictions.
Hot or not? Forecasting cellular network hot spots using sector performance indicators
Our results indicate that, compared to the best baseline, tree-based models can deliver up to 14% better forecasts for regular hot spots and 153% better forecasts for non-regular hot spots.
SEGAN: Speech Enhancement Generative Adversarial Network
In contrast to current techniques, we operate at the waveform level, training the model end-to-end, and incorporate 28 speakers and 40 different noise conditions into the same model, such that model parameters are shared across them.
A genetic algorithm to discover flexible motifs with support
Finding repeated patterns or motifs in a time series is an important unsupervised task that has still a number of open issues, starting by the definition of motif.
Ranking and significance of variable-length similarity-based time series motifs
Specifically, we find that length-normalized motif dissimilarities still have intrinsic dependencies on the motif length, and that lowest dissimilarities are particularly affected by this dependency.
Particle swarm optimization for time series motif discovery
In this article, we propose an innovative standpoint and present a solution coming from it: an anytime multimodal optimization algorithm for time series motif discovery based on particle swarms.
An Empirical Evaluation of Similarity Measures for Time Series Classification
In particular, the similarity measure is the most essential ingredient of time series clustering and classification systems.
