Unified detection of isolated and overlapping audio events using CNN-RNN.
problem Detecting both isolated and overlapping audio events simultaneously.
method Multi-label multi-task framework based on CNN-RNN, with sequential losses.
result Good generalization on isolated and overlapping audio event detection datasets.
Compact audio event descriptor from regressor confidence scores.
problem Efficiently representing audio events for classification.
method Class-specific regressors trained with random forests to estimate event onset and offset.
result Simple linear classifiers perform better than state-of-the-art on audio event classification.
Transformer model estimates keywords for better audio captioning.
problem Indeterminacy in word selection for audio events/scenes.
method Transformer-based model with keyword estimation.
result Achieved state-of-the-art performance in AAC.
Self-supervised attention model improves weakly labeled audio event classification.
problem Efficiently classify audio events with minimal labeled data.
method Develops a self-supervised attention model for weakly labeled audio clips.
result Self-supervised attention model performs comparably to strongly supervised model trained with strong labels.
This paper analyzes sound event detection in synthetic office audio, comparing different systems.
problem Comparing sound event detection systems in synthetic office audio.
method Analysis of systems submitted to DCASE 2016 task, using synthetic office sounds.
result Statistical analysis of results, highlighting system performance under controlled conditions.
FSD50K provides an open dataset of over 51k audio clips for sound event recognition.
problem Small and domain-specific sound event recognition datasets.
method Creation of an open dataset with over 51k audio clips manually labeled using 200 classes.
result FSD50K is a new open benchmark for sound event recognition research.
Study improves radio show segmentation using audio embeddings.
problem Automated segmentation of radio shows.
method Created audio embeddings from multi-class classification tasks on different datasets, evaluated performance against text-only baseline.
result Audio embeddings from non-speech sound event classification significantly outperformed text-only baseline by 32.3% in F1-measure.
Improved audio event recognition using audiovisual transformers.
problem Weakly labeled audio event classification with limited temporal information.
method Adapted transformer architecture that combines audio and video inputs.
result State-of-the-art performance on audiovisual data set.
Method trains deep neural networks on weakly labeled audio data efficiently.
problem Limited training data and lack of temporal labels for audio event detection.
method Multi-instance learning with a new loss function for stacked CNN-RNN.
result Improved performance on low-resource audio datasets.
Meta-learning improves few-shot acoustic event detection.
problem Detecting new audio events with limited labeled data.
method Formulated few-shot AED problem; explored supervised and meta-learning approaches.
result Meta-learning achieves superior performance in few-shot AED.
Study presents a dataset and methods to handle noisy labels in sound event classification.
problem Label noise in sound event classification datasets.
method Developed a dataset with noisy labels and evaluated CNN baseline systems.
result Training with large amounts of noisy data can outperform training with carefully-labeled data.
Unsupervised learning identifies sound categories from unlabeled audio.
problem Learning sound categories from unlabeled audio recordings.
method Class-agnostic semantic constraints and triplet loss-based training.
result Low-dimensional embeddings recover 41% and 84% of supervised performance.
CNNs perform well on large audio classification datasets.
problem Classifying soundtracks of videos with 30,871 labels.
method Used various CNN architectures (DNN, AlexNet, VGG, Inception, ResNet) and embeddings for audio classification.
result A model using embeddings from CNNs outperforms raw features on the Audio Set AED task.
Proposes COALA method for learning audio representations aligned with tags.
problem Lack of annotated data for high-performance audio representation learning.
method Aligns latent representations of audio and tags using a contrastive loss.
result Audio embedding model captures both acoustic and semantic characteristics.
A method for semi-supervised sound event detection using teacher-student learning.
problem Weakly-labeled data in sound event detection.
method Guided Learning with a teacher model for audio tagging and a student model for boundary detection.
result The method improves boundary detection performance using unlabeled data.
System tackles indeterminacies in automated audio captioning.
problem Word selection and sentence length indeterminacies in automated audio captioning.
method Solves caption generation and sub-indeterminacy problems through multi-task learning to estimate keywords and sentence length.
result Model achieved 20.7 SPIDEr score, significantly outperforming baseline.
New method separates sounds with weak labels in noisy environments.
problem Training audio source separation systems with limited labeled data.
method Proposes objective functions and network architectures for weakly labeled training.
result Achieves significant SI-SDR improvement in noisy scenarios.
Paper improves sound event detection using semi-supervised learning.
problem Weakly labeled sound event detection in polyphonic audio clips.
method Combines tri-training and adversarial learning for semi-supervised learning.
result Significant performance improvement over baseline model.
Improves audio transcription on scarce data with factorized tasks.
problem Weakly labelled data and lack of training samples.
method Factorizing audio transcription into multiple tasks and training a stacked CNN-RNN model.
result Different training methods for intermediate tasks have varying advantages and disadvantages.
Reduces sound event localization error by 2.6x with hybrid parametric-deep learning.
problem Sound event localization and detection accuracy.
method Hybrid approach combining parametric spatial audio analysis and deep learning.
result Reduction of localization error by 2.6x compared to baseline.
Classifiers and beamforming algorithms improved audio surveillance detection accuracy.
problem Detecting surveillance sound events with high accuracy and efficiency.
method Evaluated seven classifiers and two beamforming algorithms; used data augmentation and tested with varying SNR levels.
result SVM and Delay-and-Sum (DaS) combination achieved the highest accuracy (86.0%), but had high computational cost.
Crowd-sourced mosquito audio dataset for malaria research.
problem Understanding mosquito locations for malaria reduction.
method Release of a large mosquito audio dataset with labels from contributors.
result Demonstrated the feasibility of training a CNN on mosquito audio data.
Synthesizes images from audio and visual data using spike-based autoencoders.
problem Extracting meaningful information from spatio-temporal data for image synthesis.
method Spike-based autoencoders trained to learn spatio-temporal representations of audio and visual data.
result Synthesized images from audio samples with high fidelity, achieving competitive performance.
Paper presents an audiovisual model to recognize sounds from weakly labeled video data.
problem Sound recognition from weakly labeled video data.
method Audiovisual fusion model with attention mechanism.
result The model achieves a mean Average Precision (mAP) of 46.16 on AudioSet, outperforming state-of-the-art models.
New dataset and models generate piano music with coherent structure across multiple timescales.
problem Generating coherent musical structure with neural networks is challenging.
method Used notes as an intermediate representation to model and synthesize music across multiple timescales.
result Trained models capable of transcribing, composing, and synthesizing audio waveforms with coherent musical structure.
Paper proposes MTL for weakly labelled SED, improving performance with 2-step attention.
problem Weakly labelled sound event detection.
method Multi-Task Learning framework with 2-step Attention Pooling.
result Improved SED performance with 22.3%, 12.8%, 5.9% gains at 0, 10, 20 dB SNR.
Smart city surveillance benefits from sound event recognition.
problem Improving monitoring capabilities in smart cities.
method Exploration of several classifiers on the SESA dataset.
result SGD achieved 72.13% accuracy in sound event recognition.
Improved acoustic scene classification with factorized CNN.
problem Acoustic scene classification in varying environments.
method Large-margin factorized CNN with triplet loss.
result Improved performance and better generalization on unseen data.
Model counts event occurrences to detect and locate events in data.
problem Training deep neural networks with precise event locations.
method Weakly-supervised learning using occurrence counts.
result Comparable performance to fully-supervised methods with weaker training data.
A system is presented that segments, clusters and predicts musical audio in an unsupervised manner, adjusting the number of (timbre) clusters instantaneously to the audio input. A sequence learning algorithm adapts its structure to a dynamically changing clustering tree. The flow of the system is as follows: 1) segment…
Machine learning detects and diagnoses coughs for respiratory infections.
problem Detecting and diagnosing coughs for respiratory infections.
method Convolutional Neural Networks (CNNs) to detect cough events and diagnose three illnesses.
result Accuracy over 89% for detection and diagnosis.
Paper proposes active learning for sound event detection with reduced annotation effort.
problem Reducing annotation effort for sound event detection.
method Change point detection for candidate selection, mismatch-first farthest-traversal for selection, training with context recordings.
result The proposed system achieves similar performance to full annotation with only 2% of data, reducing annotation effort.
Simple methods boost sound event classifier accuracy by 2.5%.
problem Label noise in sound event classification datasets.
method Label smoothing regularization, mixup, and noise-robust loss functions.
result Simple methods can boost accuracy by up to 2.5%.
ACI converts call center conversations into actionable data.
problem Real-time spoken language understanding for call center conversations.
method Combines speech recognition, entity and intent recognition, and a business rules engine.
result ACI converts live audio into structured events for real-time supervision and assistance.
The study improves pitch detection in polyphonic music by learning harmonic priors.
problem Challenges in transcribing polyphonic music due to overlapping harmonics.
method Introduced Gaussian process priors and used variational Bayes for inference.
result Learning priors that fit the frequency content of sound events improves pitch detection.
Proposes a self-supervised method for generating spatial audio from monaural audio and video.
problem Generating spatial audio from monaural audio and video recordings is challenging and expensive.
method Uses a self-supervised network with an auxiliary classifier to classify video channels and generate spatial audio.
result The proposed method effectively generates spatial audio from monaural audio and video.
Deep audio prior uses neural networks to solve audio problems without data.
problem Challenging audio problems like source separation, editing, and synthesis.
method Randomly-initialized neural network with carefully designed audio prior.
result Superior audio results on Universal-150 benchmark dataset.
End-to-end sound event detection uses learned time-frequency representations.
problem Sound event detection using standard features like mel spectrogram.
method End-to-end approach with a feedforward layer block and convolutional recurrent network.
result The learned time-frequency representations improve performance over fixed features.
Zero-shot audio classification using class label embeddings.
problem Classifying audio without labeled data.
method Bilinear model with audio feature embeddings and class label embeddings.
result Achieved accuracy up to 39.7% for natural audio categories.
A new autoencoder speeds up audio style transfer.
problem Efficiently generating stylized audio from content audio.
method Autoencoder architecture for real-time audio style transfer.
result Single forward pass for stylized audio generation.
AVEC 2019 challenges AI in detecting depression and cross-cultural emotions.
problem Detecting depression and cross-cultural emotions from audiovisual data.
method Comparison of machine learning methods under standardized conditions.
result Baseline system performance on state-of-mind, depression, and cross-cultural tasks.
New model detects crying in real-world settings with improved accuracy.
problem Generalization of cry detection models to real-world environments.
method Evaluated machine learning approaches on a novel dataset of real-world infant crying.
result Improved F1 score of 0.613 for crying event recognition in real-world settings.
MCLNN improves sound recognition by learning frequency bands.
problem Efficiently recognizing acoustic events from audio signals.
method MCLNN uses a binary mask to force sparseness in network weights, focusing on frequency bands.
result MCLNN achieves competitive performance in sound recognition compared to state-of-the-art methods.
The paper proposes a method to predict audio ad quality using acoustic features.
problem Improving user experience in online music streaming services by ensuring high quality audio advertisements.
method The paper proposes predicting audio ad quality using acoustic features and a proxy metric called Long Click Rate (LCR). A deep learning model is also introduced.
result The proposed deep learning model outperforms other models trained on hand-crafted features for audio ad quality prediction.
Method learns audio embeddings with contextualized tags.
problem Align audio and tags for cross-modal tasks.
method Audio autoencoder, word embeddings, multi-head self-attention, contrastive loss.
result Multi-head self-attention improves audio representations.
Generative model improves audio synthesis from TF features.
problem Challenges in generating high-quality audio from TF features.
method Used a GAN to generate invertible TF features from short-time Fourier transforms.
result Generated TF features from GAN outperformed direct waveform generation.
Task describes tagging audio clips from 41 categories.
problem Tagging audio clips from diverse categories.
method Baseline system using AudioSet labels.
result Demonstrates feasibility of general-purpose audio tagging.
STRODE learns timings and dynamics from unlabeled time series data.
problem Learning dynamics of random event timings from unlabeled sensory inputs.
method Probabilistic Ordinary Differential Equation (STRODE) that samples from posterior point processes.
result Successfully infers event timings from synthetic and real-world datasets.