Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Oh, Hyunseok, Yi, Juheon, Lee, Youngki
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2407.00888
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910509282885632
author Oh, Hyunseok
Yi, Juheon
Lee, Youngki
author_facet Oh, Hyunseok
Yi, Juheon
Lee, Youngki
contents Transformer-based models recently reached state-of-the-art single-channel speech separation accuracy; However, their extreme computational load makes it difficult to deploy them in resource-constrained mobile or IoT devices. We thus present Papez, a lightweight and computation-efficient single-channel speech separation model. Papez is based on three key techniques. We first replace the inter-chunk Transformer with small-sized auditory working memory. Second, we adaptively prune the input tokens that do not need further processing. Finally, we reduce the number of parameters through the recurrent transformer. Our extensive evaluation shows that Papez achieves the best resource and accuracy tradeoffs with a large margin. We publicly share our source code at \texttt{https://github.com/snuhcs/Papez}
format Preprint
id arxiv_https___arxiv_org_abs_2407_00888
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Papez: Resource-Efficient Speech Separation with Auditory Working Memory
Oh, Hyunseok
Yi, Juheon
Lee, Youngki
Sound
Computation and Language
Machine Learning
Audio and Speech Processing
Transformer-based models recently reached state-of-the-art single-channel speech separation accuracy; However, their extreme computational load makes it difficult to deploy them in resource-constrained mobile or IoT devices. We thus present Papez, a lightweight and computation-efficient single-channel speech separation model. Papez is based on three key techniques. We first replace the inter-chunk Transformer with small-sized auditory working memory. Second, we adaptively prune the input tokens that do not need further processing. Finally, we reduce the number of parameters through the recurrent transformer. Our extensive evaluation shows that Papez achieves the best resource and accuracy tradeoffs with a large margin. We publicly share our source code at \texttt{https://github.com/snuhcs/Papez}
title Papez: Resource-Efficient Speech Separation with Auditory Working Memory
topic Sound
Computation and Language
Machine Learning
Audio and Speech Processing
url https://arxiv.org/abs/2407.00888