On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,181 papers · 148 categories

Trend · papers per month

count % of corpus

8 results for “spectro-temporal”

Kalman-based Spectro-Temporal ECG Analysis using Deep Convolutional Networks for Atrial Fibrillation Detectioneess.SP

Novel ECG classification for AF using spectro-temporal Kalman filtering and deep CNN.

problem Atrial fibrillation (AF) detection in ECG signals.

method Spectro-temporal representation using Kalman filter and deep convolutional neural networks.

result Proposed method achieves an overall F1 score of 80.2% on the PhysioNet/Computing in Cardiology (CinC) 2017 dataset.

Kalman filter spectro-temporal representation deep CNN atrial fibrillation ECG classification

2018-12-12abs ↗pdf ↗

WildMix Dataset and Spectro-Temporal Transformer Model for Monoaural Audio Source Separationcs.LG

New WildMix dataset and Spectro-Temporal Transformer model for better monoaural audio source separation.

problem Challenging monoaural audio source separation.

method Introducing WildMix dataset and Spectro-Temporal Transformer model (STT) with Spectro-Temporal Encoder (STE).

result STT outperforms previous baselines on the WildMix dataset.

audio source separation wildmix dataset spectro-temporal transformer monoaural machine learning

2019-11-21abs ↗pdf ↗

Semi-supervised multichannel speech enhancement with variational autoencoders and non-negative matrix factorizationcs.SD

Enhances speech in noisy environments using neural networks and NMF.

problem Speaker-independent multichannel speech enhancement in unknown noisy conditions.

method Uses variational autoencoders for supervised speech modeling and NMF for unsupervised noise modeling.

result The proposed approach outperforms NMF-based methods in noisy environments.

speech enhancement variational autoencoders non-negative matrix factorization multichannel unknown noise

2018-11-16abs ↗pdf ↗

Improved TDNNs using Deep Kernels and Frequency Dependent Grid-RNNscs.CL

Improved TDNNs enhance speech recognition with deep kernels and frequency-dependent processing.

problem Shallow TDNN models limit long-term context modeling.

method Deepened kernels with residual connections and spectro-temporal processing.

result Deep kernel TDNNs reduce WER by 6% and further by 9% with frequency-dependent Grid-RNN.

TDNNs speech recognition deep learning residual connections spectro-temporal

2018-02-18abs ↗pdf ↗

Using recurrences in time and frequency within U-net architecture for speech enhancementcs.LG

New neural network design improves speech enhancement metrics.

problem Improving speech enhancement metrics in noisy conditions.

method Combination of convolutional and recurrent layers in U-net architecture.

result Proposed solution outperforms current state-of-the-art in SDR, SIR, and STOI metrics.

speech enhancement U-net recurrent layers neural networks signal processing

2018-11-16abs ↗pdf ↗

Exploring spectro-temporal features in end-to-end convolutional neural networkscs.LG

The paper explores modifications to filter banks for speech recognition.

problem Improving speech recognition accuracy using modified filter banks.

method The authors investigate replacing triangular filters with Gabor or Gammatone filters, and rearranging filter bank computations to integrate features over smaller time scales.

result No significant improvements in phone error rate were observed with the modifications.

speech recognition filter banks Gabor filters Gammatone filters end-to-end CNN

2019-01-01abs ↗pdf ↗

State-space deep Gaussian processes with applicationsstat.ME

Develops state-space deep Gaussian processes for irregular signals.

problem Solving deep Gaussian process regression problems for irregular signals/functions.

method Represent DGPs as SDEs, solve using state-space filtering and smoothing methods.

result Rich class of priors compatible with irregular signals/functions.

Gaussian processes state-space models SDEs filtering smoothing

2021-11-24abs ↗pdf ↗

Learning Waveform-Based Acoustic Models using Deep Variational Convolutional Neural Networksstat.ML

Deep neural network learns robust acoustic models from speech waveforms.

problem Robustness in speech recognition systems using standard feature extraction techniques.

method Deep convolutional neural network with stochastic variational inference and cosine modulated filters.

result Superior performance compared to baseline waveform-based models and deep CNNs with FBANK features.

speech recognition deep learning variational inference waveform models stochastic neural networks

2019-06-23abs ↗pdf ↗