Cart
Create Account
Sign In
Already purchased this program?
Login to View
This video program is a part of the Premium package:
Quartznet: Deep Automatic Speech Recognition With 1D Time-Channel Separable Convolutions
- IEEE MemberUS $11.00
- Society MemberUS $0.00
- IEEE Student MemberUS $11.00
- Non-IEEE MemberUS $15.00
Quartznet: Deep Automatic Speech Recognition With 1D Time-Channel Separable Convolutions
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 la
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 la
