English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Spoken Language Understanding (SLU) is a critical speech recognitionapplication and is often deployed on edge devices. Consequently, on-deviceprocessing plays a significant role in the practical implementation of SLU.This paper focuses on the end-to-end (E2E) SLU model due to its small latencyproperty, unlike a cascade system, and aims to minimize the computational cost.We reduce the model size by applying tensor decomposition to the Conformer andE-Branchformer architectures used in our E2E SLU models. We propose to applysingular value decomposition to linear layers and the Tucker decomposition toconvolution layers, respectively. We also compare COMP/PARFAC decomposition andTensor-Train decomposition to the Tucker decomposition. Since the E2E model isrepresented by a single neural network, our tensor decomposition can flexiblycontrol the number of parameters without changing feature dimensions. On theSTOP dataset, we achieved 70.9% exact match accuracy under the tight constraintof only 15 million parameters.

Tensor decomposition for minimization of E2E SLU model toward on-device processing