Saved in:
Bibliographic Details
Main Authors: Kim, Gwantae, Lee, Bokyeung, Kim, Donghyeon, Ko, Hanseok
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.02562
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913377609056256
author Kim, Gwantae
Lee, Bokyeung
Kim, Donghyeon
Ko, Hanseok
author_facet Kim, Gwantae
Lee, Bokyeung
Kim, Donghyeon
Ko, Hanseok
contents In recent times, there has been a growing interest in utilizing personalized large models on low-spec devices, such as mobile and CPU-only devices. However, utilizing a personalized large model in the on-device is inefficient, and sometimes limited due to computational cost. To tackle the problem, this paper presents the weights separation method to minimize on-device model weights using parameter-efficient fine-tuning methods. Moreover, some people speak multiple languages in an utterance, as known as code-switching, the personalized ASR model is necessary to address such cases. However, current multilingual speech recognition models are limited to recognizing a single language within each utterance. To tackle this problem, we propose code-switching speech recognition models that incorporate fine-tuned monolingual and multilingual speech recognition models. Additionally, we introduce a gated low-rank adaptation(GLoRA) for parameter-efficient fine-tuning with minimal performance degradation. Our experiments, conducted on Korean-English code-switching datasets, demonstrate that fine-tuning speech recognition models for code-switching surpasses the performance of traditional code-switching speech recognition models trained from scratch. Furthermore, GLoRA enhances parameter-efficient fine-tuning performance compared to conventional LoRA.
format Preprint
id arxiv_https___arxiv_org_abs_2406_02562
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Gated Low-rank Adaptation for personalized Code-Switching Automatic Speech Recognition on the low-spec devices
Kim, Gwantae
Lee, Bokyeung
Kim, Donghyeon
Ko, Hanseok
Audio and Speech Processing
Artificial Intelligence
Computation and Language
In recent times, there has been a growing interest in utilizing personalized large models on low-spec devices, such as mobile and CPU-only devices. However, utilizing a personalized large model in the on-device is inefficient, and sometimes limited due to computational cost. To tackle the problem, this paper presents the weights separation method to minimize on-device model weights using parameter-efficient fine-tuning methods. Moreover, some people speak multiple languages in an utterance, as known as code-switching, the personalized ASR model is necessary to address such cases. However, current multilingual speech recognition models are limited to recognizing a single language within each utterance. To tackle this problem, we propose code-switching speech recognition models that incorporate fine-tuned monolingual and multilingual speech recognition models. Additionally, we introduce a gated low-rank adaptation(GLoRA) for parameter-efficient fine-tuning with minimal performance degradation. Our experiments, conducted on Korean-English code-switching datasets, demonstrate that fine-tuning speech recognition models for code-switching surpasses the performance of traditional code-switching speech recognition models trained from scratch. Furthermore, GLoRA enhances parameter-efficient fine-tuning performance compared to conventional LoRA.
title Gated Low-rank Adaptation for personalized Code-Switching Automatic Speech Recognition on the low-spec devices
topic Audio and Speech Processing
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2406.02562