Search Results for author:
Found 14 papers, 4 papers with code
Scaling ASR Improves Zero and Few Shot Learning
With 4. 5 million hours of English speech from 10 different sources across 120 countries and models of up to 10 billion parameters, we explore the frontiers of scale for automatic speech recognition.
Contextualized Streaming End-to-End Speech Recognition with Trie-Based Deep Biasing and Shallow Fusion
How to leverage dynamic contextual information in end-to-end speech recognition has remained an active research area.
Deep Shallow Fusion for RNN-T Personalization
End-to-end models in general, and Recurrent Neural Network Transducer (RNN-T) in particular, have gained significant traction in the automatic speech recognition community in the last few years due to their simplicity, compactness, and excellent performance on generic transcription tasks.
Alignment Restricted Streaming Recurrent Neural Network Transducer
There is a growing interest in the speech community in developing Recurrent Neural Network Transducer (RNN-T) models for automatic speech recognition (ASR) applications.
Contextual RNN-T For Open Domain ASR
By using an attention model and a biasing model to leverage the contextual metadata that accompanies a video, we observe a relative improvement of about 16% in Word Error Rate on Named Entities (WER-NE) for videos with related metadata.
N-HANS: Introducing the Augsburg Neuro-Holistic Audio-eNhancement System
N-HANS is a Python toolkit for in-the-wild audio enhancement, including speech, music, and general audio denoising, separation, and selective noise or source suppression.
Sound Audio and Speech Processing
Single-Channel Speech Separation with Auxiliary Speaker Embeddings
We present a novel source separation model to decompose asingle-channel speech signal into two speech segments belonging to two different speakers.
Scaling Speech Enhancement in Unseen Environments with Noise Embeddings
We address the problem of speech enhancement generalisation to unseen environments by performing two manipulations.
Calibrated Prediction Intervals for Neural Network Regressors
Despite their obvious aforementioned advantage in relation to accuracy, contemporary neural networks can, generally, be regarded as poorly calibrated and as such do not produce reliable output probability estimates.
Weakly Supervised One-Shot Detection with Attention Similarity Networks
Neural network models that are not conditioned on class identities were shown to facilitate knowledge transfer between classes and to be well-suited for one-shot learning tasks.
The Principle of Logit Separation
In contrast, there are known loss functions, as well as novel batch loss functions that we propose, which are aligned with this principle.
Fast Single-Class Classification and the Principle of Logit Separation
Our experiments show that indeed in almost all cases, losses that are aligned with the Principle of Logit Separation obtain at least 20% relative accuracy improvement in the SLC task compared to losses that are not aligned with it, and sometimes considerably more.
Tunable Sensitivity to Large Errors in Neural Network Training
We propose incorporating this idea of tunable sensitivity for hard examples in neural network learning, using a new generalization of the cross-entropy gradient step, which can be used in place of the gradient in any gradient-based training method.
Convolutional RNN: an Enhanced Model for Extracting Features from Sequential Data
Traditional convolutional layers extract features from patches of data by applying a non-linearity on an affine function of the input.
