QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions
arXiv:1910.10261
Abstract
We propose a new end-to-end neural acoustic model for automatic speech recognition. The model is composed of multiple blocks with residual connections between them. Each block consists of one or more modules with 1D time-channel separable convolutional layers, batch normalization, and ReLU layers. It is trained with CTC loss. The proposed network achieves near state-of-the-art accuracy on LibriSpeech and Wall Street Journal, while having fewer parameters than all competing models. We also demonstrate that this model can be effectively fine-tuned on new datasets.
Submitted to ICASSP 2020
References in corpus (12)
- MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
- SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
- Instance Normalization: The Missing Ingredient for Fast Stylization
- Improved Regularization of Convolutional Neural Networks with Cutout
- Mixed Precision Training
- Rigid-Motion Scattering for Texture Classification
- NeMo: a toolkit for building AI applications using Neural Modules
- Stochastic Gradient Methods with Layer-wise Adaptive Moments for Training of Deep Networks
- Fully Convolutional Speech Recognition
- Towards better decoding and language model integration in sequence to sequence models
- Jasper: An End-to-End Convolutional Neural Acoustic Model
- Letter-Based Speech Recognition with Gated ConvNets
Cited by in corpus (5)
- Conformer: Convolution-augmented Transformer for Speech Recognition
- ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Context
- MatchboxNet: 1D Time-Channel Separable Convolutional Neural Network Architecture for Speech Commands Recognition
- Improving Noise Robustness of an End-to-End Neural Model for Automatic Speech Recognition
- ConVoice: Real-Time Zero-Shot Voice Style Transfer with Convolutional Network