mlsanityData to 7 Oct 2026

Trainable Frontend For Robust and Far-Field Keyword Spotting

arXiv 1607.05666 ↗First posted 19 Jul 2016cs.CL cs.NEcited by 32

Not on the map. Why?

Abstract (arXiv)

Robust and far-field speech recognition is critical to enable true hands-free communication. In far-field conditions, signals are attenuated due to distance. To improve robustness to loudness variation, we introduce a novel frontend called per-channel energy normalization (PCEN). The key ingredient of PCEN is the use of an automatic gain control based dynamic compression to replace the widely used static (such as log or root) compression. We evaluate PCEN on the keyword spotting task. On our large rerecorded noisy and far-field eval sets, we show that PCEN significantly improves recognition performance. Furthermore, we model PCEN as neural network layers and optimize high-dimensional PCEN parameters jointly with the keyword spotting acoustic model. The trained PCEN frontend demonstrates significant further improvements without increasing model complexity or inference-time cost.

Key facts. Cited by 32 papers in the map (data through 7 Oct 2026). First posted on arXiv 19 Jul 2016.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 9; 2022: 9; 2023: 4; 2024: 5; 2025: 0; 2026: 5.

32 papers.

Sort:

  1. 2021LEAF: A Learnable Frontend for Audio Classificationcited by 69
  2. 2021Deep Spoken Keyword Spotting: An Overviewcited by 43
  3. 2023BIRB: A Generalization Benchmark for Information Retrieval in Bioacousticscited by 19
  4. 2021Real-time Denoising and Dereverberation with Tiny Recurrent U-Netcited by 16
  5. 2022Domain Generalization with Relaxed Instance Frequency-wise Normalization for Multi-device Acoustic Scene Classificationcited by 15
  6. 2024HeAR -- Health Acoustic Representationscited by 15
  7. 2021Visual Keyword Spotting with Attentioncited by 11
  8. 2022Tiny, always-on and fragile: Bias propagation through design choices in on-device machine learning workflowscited by 8
  9. 2022EfficientLEAF: A Faster LEarnable Audio Frontend of Questionable Usecited by 8
  10. 2021Domain Generalization on Efficient Acoustic Scene Classification using Residual Normalizationcited by 4
  11. 2022Learning Temporal Resolution in Spectrogram for Audio Classificationcited by 4
  12. 2021Sound Event Detection in Urban Audio With Single and Multi-Rate PCENcited by 3
20 more
  1. 2022Segment-level Metric Learning for Few-shot Bioacoustic Event Detectioncited by 3
  2. 2021Spectrum Correction: Acoustic Scene Classification with Mismatched Recording Devicescited by 2
  3. 2021Optimized Power Normalized Cepstral Coefficients towards Robust Deep Speaker Verificationcited by 2
  4. 2022An Overview of Indian Spoken Language Recognition from Machine Learning Perspectivecited by 2
  5. 2023TACNET: Temporal Audio Source Counting Networkcited by 2
  6. 2021Parameterized Channel Normalization for Far-field Deep Speaker Verificationcited by 1
  7. 2022Learning neural audio features without supervisioncited by 1
  8. 2023Spot keywords from very noisy and mixed speechcited by 1
  9. 2023Robust Wake-Up Word Detection by Two-stage Multi-resolution Ensemblescited by 1
  10. 2022Learnable Nonlinear Compression for Robust Speaker Verificationcited by 0
  11. 2022Deep Learning-Based Acoustic Mosquito Detection in Noisy Conditions Using Trainable Kernels and Augmentationscited by 0
  12. 2024Studying the Effect of Audio Filters in Pre-Trained Models for Environmental Sound Classificationcited by 0
  13. 2024Biomimetic Frontend for Differentiable Audio Processingcited by 0
  14. 2024Unsupervised Domain Adaptation Via Data Pruningcited by 0
  15. 2024Improving Distribution Alignment with Diversity-based Samplingcited by 0
  16. 2026EdgeSpot: Efficient and High-Performance Few-Shot Model for Keyword Spottingcited by 0
  17. 2026Spectrogram features for audio and speech analysiscited by 0
  18. 2026Cross-Cultural Bias in Mel-Scale Representations: Evidence and Alternatives from Speech and Musiccited by 0
  19. 2026Time-frequency localization of bird calls in dense soundscapescited by 0
  20. 2026PolyChirp: Multi-Species Birdsong Classification Using TinyML on Low-Power Acoustic Sensorscited by 0