Work

A Parameter-Efficient Learning Approach to Arabic Dialect Identification with Pre-Trained General-Purpose Speech Model

Year: 2023

Type: conference-paper

Abstract: In this work, we explore Parameter-Efficient-Learning (PEL) techniques to repurpose a General-Purpose-Speech (GSM) model for Arabic dialect identification (ADI). Specifically, we investigate different setups to incorporate trainable features into a multi-layer encoder-decoder GSM formulation under frozen pre-trained settings. Our architecture includes residual adapter and model reprogramming (input-prompting). We design a token-level label mapping to condition the GSM for Arabic Dialect Identification (ADI). This is challenging due to the high variation in vocabulary and pronunciation among the numerous regional dialects. We achieve new state-of-the-art accuracy on the ADI-17 dataset by vanilla fine-tuning. We further reduce the training budgets with the PEL method, which performs within 1.86% accuracy to fine-tuning using only 2.5% of (extra) network trainable parameters. Our study demonstrates how to identify Arabic dialects using a small dataset and limited computation with open source code and pre-trained models.

Source: Interspeech (USB)

DOI: 10.21437/interspeech.2023-1407

Authors: Srijith Radhakrishnan, Chao-Han Huck Yang, Sumeer Ahmad Khan, Narsis A. Kiani, David Gomez-Cabrero, Jesper N. Tegner

Institutions: Manipal Academy of Higher Education, King Abdullah University of Science and Technology, Georgia Institute of Technology, Karolinska Institutet

Language: English

Topic: Speech Recognition and Synthesis

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: King Abdullah University of Science and Technology

OpenAlex ID: W4385823097

FWCI: 2.331

Cites: 35

Cited by: 13

Related to: 10