QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions
Paper
•
1910.10261
•
Published
Ginsburg
borisgin
·
AI & ML interests
None yet
Organizations
authored
9
papers
over 1 year ago
Citrinet: Closing the Gap between Non-Autoregressive and Autoregressive End-to-End Models for Automatic Speech Recognition
Paper
•
2104.01721
•
Published
Hi-Fi Multi-Speaker English TTS Dataset
Paper
•
2104.01497
•
Published
TalkNet 2: Non-Autoregressive Depth-Wise Separable Convolutional Model for Speech Synthesis with Explicit Pitch and Duration Prediction
Paper
•
2104.08189
•
Published
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition
Paper
•
2305.05084
•
Published
Efficient Sequence Transduction by Jointly Predicting Tokens and Durations
Paper
•
2304.06795
•
Published
Mixed Precision Training
Paper
•
1710.03740
•
Published
TitaNet: Neural Model for speaker representation with 1D Depth-wise separable convolutions and global context
Paper
•
2110.04410
•
Published
Stochastic Gradient Methods with Layer-wise Adaptive Moments for Training of Deep Networks
Paper
•
1905.11286
•
Published