EFFUSE: Efficient Self-Supervised Feature Fusion for E2E ASR in Low Resource and Multilingual Scenarios

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Srivastava, Tejes, Shi, Jiatong, Chen, William, Watanabe, Shinji
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914826090971136
author Srivastava, Tejes
Shi, Jiatong
Chen, William
Watanabe, Shinji
author_facet Srivastava, Tejes
Shi, Jiatong
Chen, William
Watanabe, Shinji
contents Self-Supervised Learning (SSL) models have demonstrated exceptional performance in various speech tasks, particularly in low-resource and multilingual domains. Recent works show that fusing diverse SSL models could achieve superior performance compared to using one SSL model. However, fusing models increases the overall parameter size, leading to higher computational costs. We propose EFFUSE, a novel approach that uses a single SSL model to mimic the features of multiple SSL models via prediction, resulting in a lightweight framework with competitive performance. Our experiments show that EFFUSE outperforms individual SSL models in multilingual speech recognition tasks. Our best performing model achieves an average SUPERB score increase of 63.5 (6.3%) from the SSL baselines in Multilingual Speech Universal PERformance Benchmark (ML-SUPERB), while decreasing parameter size on average by 317M parameters (49%) from the fusion models.
format Preprint
id arxiv_https___arxiv_org_abs_2310_03938
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle EFFUSE: Efficient Self-Supervised Feature Fusion for E2E ASR in Low Resource and Multilingual Scenarios
Srivastava, Tejes
Shi, Jiatong
Chen, William
Watanabe, Shinji
Sound
Audio and Speech Processing
Self-Supervised Learning (SSL) models have demonstrated exceptional performance in various speech tasks, particularly in low-resource and multilingual domains. Recent works show that fusing diverse SSL models could achieve superior performance compared to using one SSL model. However, fusing models increases the overall parameter size, leading to higher computational costs. We propose EFFUSE, a novel approach that uses a single SSL model to mimic the features of multiple SSL models via prediction, resulting in a lightweight framework with competitive performance. Our experiments show that EFFUSE outperforms individual SSL models in multilingual speech recognition tasks. Our best performing model achieves an average SUPERB score increase of 63.5 (6.3%) from the SSL baselines in Multilingual Speech Universal PERformance Benchmark (ML-SUPERB), while decreasing parameter size on average by 317M parameters (49%) from the fusion models.
title EFFUSE: Efficient Self-Supervised Feature Fusion for E2E ASR in Low Resource and Multilingual Scenarios
topic Sound
Audio and Speech Processing
url https://arxiv.org/abs/2310.03938