Citrinet: Closing the Gap between Non-Autoregressive and Autoregressive End-to-End Models for Automatic Speech Recognition

Publication image

We propose Citrinet - a new end-to-end convolutional Connectionist Temporal Classification (CTC) based automatic speech recognition (ASR) model. Citrinet is deep residual neural model which uses 1D time-channel separable convolutions combined with sub-word encoding and squeeze-and-excitation. The resulting architecture significantly reduces the gap between non-autoregressive and sequence-to-sequence and transducer models. We evaluate Citrinet on LibriSpeech, TED-LIUM2, AISHELL-1 and Multilingual LibriSpeech (MLS) English speech datasets. Citrinet accuracy on these datasets is close to the best autoregressive Transducer models.


Somshubra Majumdar (NVIDIA)
Jagadeesh Balam (NVIDIA)
Oleksii Hrinchuk (NVIDIA)
Vitaly Lavrukhin (NVIDIA)
Vahid Noroozi (NVIDIA)
Boris Ginsburg (NVIDIA)

Publication Date

Research Area