Search Results for author:
Found 15 papers, 2 papers with code
SummaryMixing: A Linear-Complexity Alternative to Self-Attention for Speech Recognition and Understanding
Unfortunately, token mixing with self-attention takes quadratic time in the length of the speech utterance, slowing down inference as well as training and increasing memory consumption.
Cross-Attention is all you need: Real-Time Streaming Transformers for Personalised Speech Enhancement
Personalised speech enhancement (PSE), which extracts only the speech of a target user and removes everything else from a recorded audio clip, can potentially improve users' experiences of audio AI modules deployed in the wild.
Defensive Tensorization
We propose defensive tensorization, an adversarial defence technique that leverages a latent high-order factorization of the network.
Conditioning Sequence-to-sequence Networks with Learned Activations
Conditional neural networks play an important role in a number of sequence-to-sequence modeling tasks, including personalized sound enhancement (PSE), speaker dependent automatic speech recognition (ASR), and generative modeling such as text-to-speech synthesis.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+3
Monotonicity of the over-rotation intervals for bimodal maps
We show that the connectedness of the set of parameters for which the over-rotation interval of a bimodal interval map is constant.
Dynamical Systems
Fermionic Bell violation in the presence of background electromagnetic fields in the cosmological de Sitter spacetime
Using the squeezed state expansion, we then demonstrate the Bell violations for the vacuum and some maximally entangled initial states.
High Energy Physics - Theory General Relativity and Quantum Cosmology
Unicritical Laminations
In the end we verify the \emph{Fatou conjecture} for the unicritical laminations and extend the \emph{Lavaurs algorithm} onto $\mathrm{UML}_d$.
Dynamical Systems Primary: 54F20, Secondary: 30C35
NAS-Bench-ASR: Reproducible Neural Architecture Search for Speech Recognition
These datasets, however, focus predominantly on computer vision and NLP tasks and thus suffer from the problem of limited coverage of application domains.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+2
Bunched LPCNet : Vocoder for Low-cost Neural Text-To-Speech Systems
LPCNet is an efficient vocoder that combines linear prediction and deep neural network modules to keep the computational complexity low.
Iterative Compression of End-to-End ASR Model using AutoML
Increasing demand for on-device Automatic Speech Recognition (ASR) systems has resulted in renewed interests in developing automatic model compression techniques.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+3
Defensive Tensorization: Randomized Tensor Parametrization for Robust Neural Networks
As deep neural networks become widely adopted for solving most problems in computer vision and audio-understanding, there are rising concerns about their potential vulnerability.
MobiSR: Efficient On-Device Super-Resolution through Heterogeneous Mobile Processors
In recent years, convolutional networks have demonstrated unprecedented performance in the image restoration task of super-resolution (SR).
Understanding Opportunities for Efficiency in Single-image Super Resolution Networks
A successful application of convolutional architectures is to increase the resolution of single low-resolution images -- a image restoration task called super-resolution (SR).
Cross-modal Recurrent Models for Weight Objective Prediction from Multimodal Time-series Data
We analyse multimodal time-series data corresponding to weight, sleep and steps measurements.
Towards Using Unlabeled Data in a Sparse-coding Framework for Human Activity Recognition
We propose a sparse-coding framework for activity recognition in ubiquitous and mobile computing that alleviates two fundamental problems of current supervised learning approaches.
