Double Linear Transformer for Background Music Generation from Videos

Many music generation research works have achieved effective performance, while rarely combining music with given videos. We propose a model with two linear Transformers to generate background music according to a given video. To enhance the melodic quality of the generated music, we firstly input n...

Full description

Bibliographic Details
Main Authors: Xueting Yang, Ying Yu, Xiaoyu Wu
Format: Article
Language:English
Published: MDPI AG 2022-05-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/12/10/5050
_version_ 1827670534252396544
author Xueting Yang
Ying Yu
Xiaoyu Wu
author_facet Xueting Yang
Ying Yu
Xiaoyu Wu
author_sort Xueting Yang
collection DOAJ
description Many music generation research works have achieved effective performance, while rarely combining music with given videos. We propose a model with two linear Transformers to generate background music according to a given video. To enhance the melodic quality of the generated music, we firstly input note-related and rhythm-related music features separately into each Transformer network. In particular, we pay attention to the connection and the independence of music features. Then, in order to generate the music that matches the given video, the current state-of-the-art cross-modal inference method is set up to establish the relationship between visual mode and sound mode. Subjective and objective experiment indicate that the generated background music matches the video well and is also melodious.
first_indexed 2024-03-10T03:23:25Z
format Article
id doaj.art-6f88e849375e400da0dd0440092eeeb9
institution Directory Open Access Journal
issn 2076-3417
language English
last_indexed 2024-03-10T03:23:25Z
publishDate 2022-05-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj.art-6f88e849375e400da0dd0440092eeeb92023-11-23T09:57:00ZengMDPI AGApplied Sciences2076-34172022-05-011210505010.3390/app12105050Double Linear Transformer for Background Music Generation from VideosXueting Yang0Ying Yu1Xiaoyu Wu2Faculty of Information and Communication Engineering, Communication University of China, Beijing 100024, ChinaFaculty of Information and Communication Engineering, Communication University of China, Beijing 100024, ChinaFaculty of Information and Communication Engineering, Communication University of China, Beijing 100024, ChinaMany music generation research works have achieved effective performance, while rarely combining music with given videos. We propose a model with two linear Transformers to generate background music according to a given video. To enhance the melodic quality of the generated music, we firstly input note-related and rhythm-related music features separately into each Transformer network. In particular, we pay attention to the connection and the independence of music features. Then, in order to generate the music that matches the given video, the current state-of-the-art cross-modal inference method is set up to establish the relationship between visual mode and sound mode. Subjective and objective experiment indicate that the generated background music matches the video well and is also melodious.https://www.mdpi.com/2076-3417/12/10/5050video background music generationmusic feature extractionlinear Transformer
spellingShingle Xueting Yang
Ying Yu
Xiaoyu Wu
Double Linear Transformer for Background Music Generation from Videos
Applied Sciences
video background music generation
music feature extraction
linear Transformer
title Double Linear Transformer for Background Music Generation from Videos
title_full Double Linear Transformer for Background Music Generation from Videos
title_fullStr Double Linear Transformer for Background Music Generation from Videos
title_full_unstemmed Double Linear Transformer for Background Music Generation from Videos
title_short Double Linear Transformer for Background Music Generation from Videos
title_sort double linear transformer for background music generation from videos
topic video background music generation
music feature extraction
linear Transformer
url https://www.mdpi.com/2076-3417/12/10/5050
work_keys_str_mv AT xuetingyang doublelineartransformerforbackgroundmusicgenerationfromvideos
AT yingyu doublelineartransformerforbackgroundmusicgenerationfromvideos
AT xiaoyuwu doublelineartransformerforbackgroundmusicgenerationfromvideos