AnySimLite: A Lightweight Few-Shot Similarity Encoder for On-Device Speech-Adjacent Classification

arXiv:2606.26452 · cs.CL, cs.SD · Submitted 2026-06-24 · Read on arXiv

cs.CL, cs.SD

Submitted: 2026-06-24

Updated: 2026-06-24

Comments: Accepted at Interspeech 2026

Journal ref: Proc. Interspeech 2026, 6363-6367

DOI: 10.21437/Interspeech.2026-1316

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

The gist: To minimize privacy concerns and inference latency on edge devices like smartphones, lightweight on-device models remain important for end-user applications.

Terminology

Abstract

To minimize privacy concerns and inference latency on edge devices like smartphones, lightweight on-device models remain important for end-user applications. Many of these applications involve natural language classification, but deploying multiple specialized models creates a memory footprint challenge. We investigate: Can a single lightweight architecture solve multiple Speech-Adjacent (SA) classification tasks through reduction to a nuanced text similarity formulation? We propose AnySimLite, a lightweight similarity encoder that combines word-level and character-level channels. Together with a dataset transformation strategy, we evaluate AnySimLite across multiple SA classification tasks and show that it consistently achieves state-of-the-art (SOTA) or SOTA-competitive performance in few-shot settings while maintaining a low memory footprint. Even in the worst case, the performance drop remains below 7% while using < 1 over 250 th of the model size of the SOTA qLLaMA LoRA-7B baseline.

Sources

Related papers