QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions

Samuel Kriman, Stanislav Beliaev, Boris Ginsburg, Jocelyn Huang, Oleksii Kuchaiev +4 more
10/22/2019
eess.AS

Official code: NVIDIA-NeMo/Speech (18,606 stars, Apache-2.0, last commit 2026-10-20).

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.

View on arXivView PDF

Code Implementations(1)

NVIDIA-NeMo/SpeechOfficial100%
Apache-2.0

A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)

18,6063,656Jinja, ShellAug 5, 2019yesterdayApache-2.0
asrdeeplearninggenerative-aimachine-translationneural-networks+5 more

Cite this paper

@article{kriman2019quartznet,
  title  = {QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions},
  author = {Samuel Kriman and Stanislav Beliaev and Boris Ginsburg and Jocelyn Huang and Oleksii Kuchaiev and Vitaly Lavrukhin and Ryan Leary and Jason Li and Yang Zhang},
  year   = {2019},
  eprint = {1910.10261},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1910.10261}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1910.10261.svg)](https://codeofpaper.com/paper/1910.10261)

Discussion