Multi-Scale Approaches to the MediaEval 2015 "Emotion in Music" Task

نویسندگان

  • Mingxing Xu
  • Xinxing Li
  • Haishu Xianyu
  • Jiashen Tian
  • Fanhang Meng
  • Wenxiao Chen
چکیده

The goal of the “Emotion in Music” task in MediaEval 2015 is to automatically estimate the emotions expressed by music (in terms of Arousal and Valence) in a time-continuous fashion. In this paper, considering the high context correlation among the music feature sequence, we study several multiscale approaches at different levels, including acoustic feature learning with Deep Brief Networks (DBNs) followed a modified Autoencoder (AE), bi-directional Long-Short Term Memory Recurrent Neural Networks (BLSTM-RNNs) based multi-scale regression fusion with Extreme Learning Machine (ELM), and hierarchical prediction with Support Vector Regression (SVR). The evaluation performances of all runs submitted are significantly better than the baseline provided by the organizers, illustrating the effectiveness of the proposed approaches.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

PKU-AIPL' Solution for MediaEval 2015 Emotion in Music Task

In this paper, we describe the PKU-AIPL Team solution of Emotion in Music task in MediaEval benchmarking campaign 2015. We extracted and designed several sets of features and used continuous conditional random field(CCRF) for dynamic emotion characterization task.

متن کامل

MediaEval 2015: JKU-Tinnitus Approach to Emotion in Music Task

This paper describes the JKU-Tinnitus submission to the “Emotion in Music” task [1] of the 2015 MediaEval Benchmark. Given a set of manually annotated music and a set of features for each music file, machine learning algorithms are applied to estimate the development of emotional arousal and valence over the course of a piece of music. Our pipeline roughly contains feature extraction from the m...

متن کامل

MediaEval 2014: THU-HCSIL Approach to Emotion in Music Task using Multi-level Regression

This working notes paper describes the system proposed by THU-HCSIL team for dynamic music emotion recognition. The procedure is divided into two module feature extraction and regression. Both feature selection and feature combination are used to form the final THU feature set. In regression module, a Booster-based Multi-level Regression method is presented, which outperforms the baseline signi...

متن کامل

Emotion in Music Task at MediaEval 2015

The Emotion in Music task is held for the third consecutive year at the MediaEval benchmarking campaign. The unceasing interest towards the task shows that the music emotion recognition (MER) problem is truly important to the community, and there is a lot remaining to be discovered about it. Automatic MER methods could greatly improve the accessibility of music collections by providing quick an...

متن کامل

Emotion in Music task: Lessons Learned

The Emotion in Music task was organized within MediaEval benchmarking campaign during three consecutive years, from 2013 to 2015. In this paper we describe the challenges we faced and the solutions we found. We used crowdsourcing on Amazon Mechanical Turk to annotate a corpus of music pieces with continuous (per-second) emotion annotations. To assure sufficient quality of the data, the annotati...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2015