Spoken Language Understanding on the Edge

We consider the problem of performing Spoken Language Understanding (SLU) on small devices typical of IoT applications. Our contribution is two-fold. First, we outline the design of an embedded, private-by-design SLU system and show that it has performance on-par with cloud-based commercial solutions. Second, we release the datasets used in our experiments in the interest of reproducibility and in the hope that they can prove useful to the community.

cs.CL cs.LG cs.SD eess.AS Computer Science Engineering

View graph Source

Paper

References (22)

01Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfacesA. Coucke, Alaa Saade, Adrien Ball et al.2018 · arXiv.org · 925 citations In Library

02Speech Commands: A Dataset for Limited-Vocabulary Speech RecognitionPete Warden2018 · arXiv.org · 2.1k citations In Library

03Achieving Human Parity in Conversational Speech RecognitionWayne Xiong, J. Droppo, Xuedong Huang et al.2016 · arXiv.org · 594 citations In Library

04How to Add Word Classes to the Kaldi Speech Recognition ToolkitAxel Horndasch, Caroline Kaufhold, Elmar Nöth2016 · International Conference on Text, Speech and Dialogue · 11 citations

05Purely Sequence-Trained Neural Networks for ASR Based on Lattice-Free MMIDaniel Povey, Vijayaditya Peddinti, Daniel Galvez et al.2016 · Interspeech · 891 citations

06Personalized speech recognition on mobile devicesIan McGraw, Rohit Prabhavalkar, R. Álvarez et al.2016 · IEEE International Conference on Acoustics, Speech, and Signal Processing · 201 citations In Library

07Librispeech: An ASR corpus based on public domain audio booksVassil Panayotov, Guoguo Chen, Daniel Povey et al.2015 · IEEE International Conference on Acoustics, Speech, and Signal Processing · 8.2k citations

08Using Recurrent Neural Networks for Slot Filling in Spoken Language UnderstandingGrégoire Mesnil, Yann Dauphin, K. Yao et al.2015 · IEEE/ACM Transactions on Audio Speech and Language Processing · 587 citations

09Minimum Bayes Risk decoding and system combination based on a recursion for edit distanceHaihua Xu, Daniel Povey, L. Mangu et al.2011 · Computer Speech and Language · 161 citations

10Filters for Efficient Composition of Weighted Finite-State TransducersCyril Allauzen, M. Riley, J. Schalkwyk2010 · International Conference on Implementation and Application of Automata · 20 citations

11Is word error rate a good indicator for spoken language understanding accuracyYe-Yi Wang, A. Acero, Ciprian Chelba2003 · 2003 IEEE Workshop on Automatic Speech Recognition and Understanding (IEEE Cat. No.03EX721) · 205 citations

12Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence DataJ. Lafferty, A. McCallum, Fernando Pereira2001 · International Conference on Machine Learning · 16k citations

Scroll for more · 10 remaining

Similar papers

Speech Understanding on Tiny Devices with A Learning CacheAfsara Benazir, Zhiming Xu, Felix Xiaozhu Lin2024 In Library

From Cloud to Edge: Enabling Offline IoT With Small-Scale Language ModelsAmit Sangwan, Priyangshu Sen, Arjun Singh et al.2025 · IEEE Internet of Things Magazine · 3 citations In Library

TINYS2I: A Small-Footprint Utterance Classification Model with Contextual Support for On-Device SLUAnastasios Alexandridis, Kanthashree Mysore Sathyendra, Grant P. Strimel et al.2022 · ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · 2 citations In Library

Privacy-Preserving End-to-End Spoken Language UnderstandingYinggui Wang, Wei Huang, Le Yang2023 In Library

Privacy-Preserving End-to-End Spoken Language UnderstandingYinggui Wang, Wei Huang, Le Yang2024 · arXiv (Cornell University) · 4 citations In Library

Voice to Action: Spoken Language Understanding for Memory-Constrained SystemsAshutosh Gupta, Aditya Jayasimha, Aman Maghan et al.2021 · 2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU) In Library

Speech Understanding on Tiny Devices with A Learning CacheAfsara Benazir, Zhiming Xu, Felix Xiaozhu Lin2023 · arXiv (Cornell University) In Library

Edge AI: Small Language Models on the GoJoachim J. Włodarz2026 In Library

ICASSP 2023 Spoken Language Understanding Grand ChallengeAkshat Shrivastava, Suyoun Kim, Paden Tomasello et al.2023 · IEEE International Conference on Acoustics, Speech, and Signal Processing · 1 citation In Library

Speak or Chat with Me: End-to-End Spoken Language Understanding System with Flexible InputsS. Cha, Wang Hou, Hyun Jung et al.2021 · Interspeech · 15 citations In Library

Source of record

arXiv

44B

Welcome.

This panel sticks with you. Pick anything below and it opens right beside it, so you can dig through 60,000-plus records without ever losing your spot here.

Travel 44B

LibraryPapers, policy, standards, statute LabsEvery organization building AI ModelsIntelligence, price, and speed BenchmarksThe evaluation catalog IncidentsAI failures and harms SearchOne field across everything DashboardThe state of AI accountability in NY 44B RegistryThe Art. 44-B compliance portal