music-4

Melody Extraction

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in melody-extraction-4

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find melody-extraction-4 models and implementations

Datasets

ErhuPT

Subtasks

No subtasks available.

Most implemented papers

A Streamlined Encoder/decoder Architecture for Melody Extraction

Yi-Hsuan Yang, Li Su, Tsung-Han Hsieh•Mon Oct 29 2018

A novel streamlined encoder/decoder network that is designed for melody extraction in polyphonic musical audio can achieve result close to the state-of-the-art with much fewer convolutional layers and simpler convolution modules.

57

Content

0

Paper Graph

Musical Instrument Playing Technique Detection Based on FCN: Using Chinese Bowed-Stringed Instrument as an Example

Jingru Li, Xiaoou Chen, Zijin Li, Shicheng Zhang, Baoqiang Han, Deshun Yang•Sat Oct 19 2019

An algorithm to detect on variable length audio based on Sound Event Detection by FCN for musical instrument playing technique detection and chooses Erhu, a well-known Chinese bowed-stringed instrument, to experiment with the method.

5 0

Paper Graph

BERT-like Pre-training for Symbolic Piano Music Classification Tasks

Yi-Hsuan Yang, Joann Ching, Yi-Hui Chou, I-Chun Chen, Chin-Jui Chang•Sun Jul 11 2021

This article presents a benchmark study of symbolic piano music classification using the masked language modelling approach of the Bidirectional Encoder Representations from Transformers (BERT), and shows that the BERT approach leads to higher classification accuracy than recurrent neural network (RNN)-based baselines.

28 0

Paper Graph

Tonet: Tone-Octave Network for Singing Melody Extraction from Polyphonic Music

Taylor Berg-Kirkpatrick, S. Dubnov, Ke Chen, Shuai Yu, Cheng-i Wang, Wei Li•Tue Feb 01 2022

TONet1, a plug-and-play model that improves both tone and octave perceptions by leveraging a novel input representation and a novel network architecture, is proposed and results show that tone-octave fusion with Tone-CFP can significantly improve the singing voice extraction performance across various datasets.

21 0

Paper Graph

Clarinet: A Music Retrieval System

Kshitij Alwadhi, Rohan Sharma, Siddhant Sharma•Sat Oct 22 2022

A MIDI based approach for music recognition is proposed and implemented, and a novel melody extraction algorithm is designed that improves recall results by more than 10%.

0 0

Paper Graph

Towards Improving Harmonic Sensitivity and Prediction Stability for Singing Melody Extraction

K. Chen, Taylor Berg-Kirkpatrick, S. Dubnov, Keren Shao•Thu Aug 03 2023

This paper proposes an input feature modification and a training objective modification based on two assumptions of harmonics in the spectrograms of audio data decay rapidly along the frequency axis to enhance the model's sensitivity on the trailing harmonics.

3 0

Paper Graph

Adding a benchmark result helps the community track progress.