{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/a-streamlined-encoderdecoder-architecture-for","title":"A Streamlined Encoder/Decoder Architecture for Melody Extraction","arxiv_id":"1810.12947","date":"2018-10-30","proceeding":null,"authors":["Tsung-Han Hsieh","Li Su","Yi-Hsuan Yang"],"abstract":"Melody extraction in polyphonic musical audio is important for music signal\nprocessing. In this paper, we propose a novel streamlined encoder/decoder\nnetwork that is designed for the task. We make two technical contributions.\nFirst, drawing inspiration from a state-of-the-art model for semantic\npixel-wise segmentation, we pass through the pooling indices between pooling\nand un-pooling layers to localize the melody in frequency. We can achieve\nresult close to the state-of-the-art with much fewer convolutional layers and\nsimpler convolution modules. Second, we propose a way to use the bottleneck\nlayer of the network to estimate the existence of a melody line for each time\nframe, and make it possible to use a simple argmax function instead of ad-hoc\nthresholding to get the final estimation of the melody line. Our experiments on\nboth vocal melody extraction and general melody extraction validate the\neffectiveness of the proposed model.","url_abs":"http://arxiv.org/abs/1810.12947v1","url_pdf":"http://arxiv.org/pdf/1810.12947v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"a-streamlined-encoderdecoder-architecture-for","repo_url":"https://github.com/bill317996/Melody-extraction-with-melodic-segnet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"melody-extraction","task_name":"Melody Extraction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}