Specialized embedding approximation for edge intelligence: A case study in urban sound classification

Sangeeta Srivastava, Dhrubojyoti Roy, Mark Cartwright, Juan P. Bello, Anish Arora

Research output: Contribution to journalConference articlepeer-review

5 Scopus citations

Abstract

Embedding models that encode semantic information into lowdimensional vector representations are useful in various machine learning tasks with limited training data. However, these models are typically too large to support inference in small edge devices, which motivates training of smaller yet comparably predictive student embedding models through knowledge distillation (KD). While knowledge distillation traditionally uses the teacher's original training dataset to train the student, we hypothesize that using a dataset similar to the student's target domain allows for better compression and training efficiency for the said domain, at the cost of reduced generality across other (non-pertinent) domains. Hence, we introduce Specialized Embedding Approximation (SEA) to train a student featurizer to approximate the teacher's embedding manifold for a given target domain. We demonstrate the feasibility of SEA in the context of acoustic event classification for urban noise monitoring and show that leveraging a dataset related to this target domain not only improves the baseline performance of the original embedding model but also yields competitive students with >1 order of magnitude lesser storage and activation memory. We further investigate the impact of using random and informed sampling techniques for dimensionality reduction in SEA.

Original languageEnglish (US)
Pages (from-to)8378-8382
Number of pages5
JournalICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2021-June
DOIs
StatePublished - 2021
Externally publishedYes
Event2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021 - Virtual, Toronto, Canada
Duration: Jun 6 2021Jun 11 2021

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Keywords

  • Acoustic event detection
  • Deep audio embeddings
  • Knowledge distillation
  • On-device machine learning
  • Urban noise classification

Fingerprint

Dive into the research topics of 'Specialized embedding approximation for edge intelligence: A case study in urban sound classification'. Together they form a unique fingerprint.

Cite this