Honk: A PyTorch Reimplementation of Convolutional Neural Networks for Keyword Spotting

Raphael Tang, Jimmy Lin
10/18/2017
cs.CL

Official code: castorini/honk (527 stars, MIT, last commit 2023-05-22).

Abstract

We describe Honk, an open-source PyTorch reimplementation of convolutional neural networks for keyword spotting that are included as examples in TensorFlow. These models are useful for recognizing "command triggers" in speech-based interfaces (e.g., "Hey Siri"), which serve as explicit cues for audio recordings of utterances that are sent to the cloud for full speech recognition. Evaluation on Google's recently released Speech Commands Dataset shows that our reimplementation is comparable in accuracy and provides a starting point for future work on the keyword spotting task.

View on arXivView PDF

Code Implementations(1)

castorini/honkOfficial100%
MIT

PyTorch implementations of neural network models for keyword spotting

527120Shell, PythonSep 14, 20173 years agoMIT

Similar Papers

Deep Residual Learning for Small-Footprint Keyword Spotting

86% match1 repo5272017

Keyword spotting -- Detecting commands in speech using deep learning

85% match2023

AraSpot: Arabic Spoken Command Spotting

84% match1 repo32023

Keyword Transformer: A Self-Attention Model for Keyword Spotting

83% match1 repo1482021

A Multitask Training Approach to Enhance Whisper with Contextual Biasing and Open-Vocabulary Keyword Spotting

82% match2023

Keyword Spotting with Hyper-Matched Filters for Small Footprint Devices

82% match2025

Cite this paper

@article{tang2017honk,
  title  = {Honk: A PyTorch Reimplementation of Convolutional Neural Networks for Keyword Spotting},
  author = {Raphael Tang and Jimmy Lin},
  year   = {2017},
  eprint = {1710.06554},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1710.06554}
}

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/1710.06554.svg)](https://codeofpaper.com/paper/1710.06554)

Discussion