mlsanityData to 7 Oct 2026

Few-Shot Open-Set Learning for On-Device Customization of KeyWord Spotting Systems

arXiv 2306.02161 ↗First posted 3 Jun 2023cs.LGcited by 9

Not on the map. Why?

Abstract (arXiv)

A personalized KeyWord Spotting (KWS) pipeline typically requires the training of a Deep Learning model on a large set of user-defined speech utterances, preventing fast customization directly applied on-device. To fill this gap, this paper investigates few-shot learning methods for open-set KWS classification by combining a deep feature encoder with a prototype-based classifier. With user-defined keywords from 10 classes of the Google Speech Command dataset, our study reports an accuracy of up to 76% in a 10-shot scenario while the false acceptance rate of unknown data is kept to 5%. In the analyzed settings, the usage of the triplet loss to train an encoder with normalized output features performs better than the prototypical networks jointly trained with a generator of dummy unknown-class prototypes. This design is also more effective than encoders trained on a classification problem and features fewer parameters than other iso-accuracy approaches.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 19 of its 26 references that are in the map. First posted on arXiv 3 Jun 2023.

Builds on

19 of 26 references.

References · 19

  1. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  2. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 741
  3. 2015Towards Open Set Deep Networkscited by 533
  4. 2020In defence of metric learning for speaker recognitioncited by 123
  5. 2017Hello Edge: Keyword Spotting on Microcontrollerscited by 108
  6. 2020A Wholistic View of Continual Learning with Deep Neural Networks: Forgotten Lessons and the Bridge to Active and Open World Learningcited by 86
  7. 2020Few-Shot Open-Set Recognition using Meta-Learningcited by 50
  8. 2022Online Continual Learning for Embedded Devicescited by 48
  9. 2020Streaming keyword spotting on mobile devicescited by 47
  10. 2021Deep Spoken Keyword Spotting: An Overviewcited by 43
  11. 2021Few-Shot Keyword Spotting in Any Languagecited by 27
  12. 2021Query-by-Example Keyword Spotting system using Multi-head Attention and Softtriple Losscited by 25
7 more
  1. 2021Learning Efficient Representations for Keyword Spotting with Triplet Losscited by 24
  2. 2020Few-Shot Keyword Spotting With Prototypical Networkscited by 18
  3. 2020Metric Learning for Keyword Spottingcited by 9
  4. 2022Dummy Prototypical Networks for Few-Shot Open-Set Keyword Spottingcited by 7
  5. 2022Personalized Keyword Spotting through Multi-task Learningcited by 7
  6. 2022Metric Learning for User-defined Keyword Spottingcited by 7
  7. 2022QbyE-MLPMixer: Query-by-Example Open-Vocabulary Keyword Spotting using MLPMixercited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 1; 2026: 6.

9 papers.

  1. 2024Self-Learning for Personalized Keyword Spotting on Ultra-Low-Power Audio Sensorscited by 7
  2. 2024TinySV: Speaker Verification in TinyML with On-device Learningcited by 5
  3. 2026Few-Shot Open-Set Audio Classification Using Attention Information-Fused Prototypescited by 2
  4. 2025Enhancing Few-shot Keyword Spotting Performance through Pre-Trained Self-supervised Speech Modelscited by 1
  5. 2026EdgeSpot: Efficient and High-Performance Few-Shot Model for Keyword Spottingcited by 0
  6. 2026OnDA: On-device Channel Pruning for Efficient Personalized Keyword Spottingcited by 0
  7. 2026Scaling few-shot spoken word classification with generative meta-continual learningcited by 0
  8. 2026What changes after deployment? A survey on On-device Learning in TinyMLcited by 0
  9. 2026Few-Shot Open-Set Audio Classification via Transductive Prototype Refinement and Class Logit Enhancementcited by 0