Filters








25 Hits in 8.8 sec

Automatic Drum Transcription Using The Student-Teacher Learning Paradigm With Unlabeled Music Data

Chih-Wei Wu, Alexander Lerch
2017 Zenodo  
CONCLUSION This paper presents a system for Automatic Drum Transcription based on the student-teacher learning paradigm with the unlabeled music data.  ...  "Automatic drum transcription using the student-teacher learning paradigm with unlabeled music data", 18th International Society for Music Information Retrieval Conference, Suzhou, China, 2017.  ... 
doi:10.5281/zenodo.1415904 fatcat:ln3ifrvqnbcbvmjknjros7doki

From Labeled to Unlabeled Data – On the Data Challenge in Automatic Drum Transcription

Chih-Wei Wu, Alexander Lerch
2018 Zenodo  
Specifically, two paradigms that harness information from unlabeled data, namely feature learning and student-teacher learning, are applied to two major types of ADT systems.  ...  Automatic Drum Transcription (ADT), like many other music information retrieval tasks, has made progress in the past years through the integration of machine learning and audio signal processing techniques  ...  INTRODUCTION Automatic drum transcription (ADT), a sub-task of Automatic Music Transcription (AMT) [2] that concerns the extraction of drum events from music signals, witnesses a growth in data-driven  ... 
doi:10.5281/zenodo.1492446 fatcat:5ruhpskjwnfijckoua7nguytau

Semi-supervised learning using teacher-student models for vocal melody extraction

Sangeun Kum, Jing-Hua Lin, Li Su, Juhan Nam
2020 Zenodo  
The teacher model is pre-trained with labeled data and guides the student model to make identical predictions given unlabeled input in a self-training setting.  ...  The results show that the SSL method significantly increases the performance against supervised learning only and the improvement depends on the teacher-student models, the size of unlabeled data, the  ...  Wu and Lerch applied the approach to automatic drum transcription [16] .  ... 
doi:10.5281/zenodo.4245374 fatcat:bsxj3gbl4nfxnemfyx2reozed4

Semi-supervised learning using teacher-student models for vocal melody extraction [article]

Sangeun Kum, Jing-Hua Lin, Li Su, Juhan Nam
2020 arXiv   pre-print
The teacher model is pre-trained with labeled data and guides the student model to make identical predictions given unlabeled input in a self-training setting.  ...  The results show that the SSL method significantly increases the performance against supervised learning only and the improvement depends on the teacher-student models, the size of unlabeled data, the  ...  Wu and Lerch applied the approach to automatic drum transcription [16] .  ... 
arXiv:2008.06358v1 fatcat:ipl4gguzcfhr7aifchpfilx7xu

Creating DALI, a Large Dataset of Synchronized Audio, Lyrics, and Notes

Gabriel Meseguer-Brocal, Alice Cohen-Hadria, Geoffroy Peeters
2020 Transactions of the International Society for Music Information Retrieval  
Our method is motivated by active learning and the teacher-student paradigm. We establish a loop whereby dataset creation and model learning interact, benefiting each other.  ...  We progressively improve our model using the collected data. At the same time, we correct and enhance the collected data every time we update the model.  ...  Acknowledgement This research has received funding from the French National Research Agency under the contract ANR-16-CE23-0017-01 (WASABI project).  ... 
doi:10.5334/tismir.30 fatcat:f4b7y65oijboxodfovest6ry6u

MULTIMODAL ANALYSIS: Informed content estimation and audio source separation [article]

Gabriel Meseguer-Brocal
2021 arXiv   pre-print
This dissertation proposes the study of multimodal learning in the context of musical signals. Throughout, we focus on the interaction between audio signals and text information.  ...  Among the many text sources related to music that can be used (e.g. reviews, metadata, or social network feedback), we concentrate on lyrics.  ...  One of the few examples that applies it to automatic drum transcription is (Wu and Lerch, 2017) , in which the teacher labels the student dataset of drum recordings.  ... 
arXiv:2104.13276v3 fatcat:wirjfj4iwjgfteejmeujydey7u

A Comprehensive Survey on Deep Music Generation: Multi-level Representations, Algorithms, Evaluations, and Future Directions [article]

Shulei Ji, Jing Luo, Xinyu Yang
2020 arXiv   pre-print
This paper attempts to provide an overview of various composition tasks under different music generation levels, covering most of the currently popular music generation tasks using deep learning.  ...  Previous surveys have explored the network models employed in the field of automatic music generation.  ...  With the advent of deep learning, using deep learning technologies to automatically generate various contents (such as images, text, etc.) has become a hot issue.  ... 
arXiv:2011.06801v1 fatcat:cixou3d2jzertlcpb7kb5x5ery

Music information retrieval

J. Stephen Downie
2005 Annual Review of Information Science and Technology  
Welcome friends and colleagues to the 2 nd Annual International Symposium on Music Information Retrieval -ISMIR 2001.  ...  This document includes the texts of the accepted papers along with the extended abstracts of the invited talks and poster presentations.  ...  Acknowledgment We would like to thank the CWEB Technology, Inc., for sharing us the data set used in our experiments.  ... 
doi:10.1002/aris.1440370108 fatcat:5v36lrlqbjfi5fkuxw3mzjyhhe

Deep Learning and Knowledge Integration for Music Audio Analysis (Dagstuhl Seminar 22082)

Meinard Müller, Rachel Bittner, Juhan Nam
2022
Given the increasing amount of digital music, the development of computational tools that allow users to find, organize, analyze, and interact with music has become central to the research field known  ...  In the Dagstuhl Seminar 22082, we critically examined this trend, discussing the strengths and weaknesses of these approaches using music as a challenging application domain.  ...  Instead of relying on large amounts of labeled data, this technique exploits known variants and invariants of a dataset, using lots of unlabeled data.  ... 
doi:10.4230/dagrep.12.2.103 fatcat:f3mb2pumxreb7h75hj56n6rs5y

Toward Meaningful Music Education in the Middle School Music Classroom: An Action Research Project

Cade McNaughton Bonar, University, My, Kay Hartwig
2018
Furthermore, emerging technologies are enabling greater accessibility to music making and production. Students can learn, create and share music using digital technologies alone.  ...  It can present a crisis of relevance for the students it proclaims to serve, with student expectations of music and musical experiences offered often existing at considerable remove (Regelski, 2005a; Lines  ...  automatically or necessarily from musical engagement.  ... 
doi:10.25904/1912/1861 fatcat:a4iu6saeujdpjotvapyea3ckju

The role of principal component analysis in vowel acquisition research

Krisztina Zajdó, Jeannette van der Stelt, Ton G. Wempe
2004 Journal of the Acoustical Society of America  
The role of principal component analysis in vowel acquisition research Zajdo, K.; van der Stelt, J.M.; Wempe, A.G.  ...  The performance of the automatic classification compares favorably with human performance.  ...  A 2IFC paradigm was used to assess amplitude modulation ͑AM͒ detection performance for stimuli whose envelope was removed using the Hilbert transform, leaving only the fine structure.  ... 
doi:10.1121/1.4785499 fatcat:cdj5floyg5cd5ch2nwidivx4nq

On becoming and being a musician: A mixed methods study of musicianship in children and adults

Dawn Rose
2016
The data generates new hypotheses that musical learning supports and encourages flexible cognitive and behavioural skills and creativity that are further enhanced by the concomitant experience of nonverbal  ...  Researchers have suggested that changes associated with musical learning may transfer to near domains (e.g. fine motor ability) and/or far domains, such as general intelligence.  ...  The mother who was hearing impaired said of her daughter who was learning to play drums, "…And you listen all the time and you like drumming along to the music you like.  ... 
doi:10.25602/gold.00019105 fatcat:24fp6aqnxvecxewbdo56wt26eu

Music Encoding Conference Proceedings 26-29 May, 2020 Tufts University, Boston (USA). Edited by Elsa De Luca and Julia Flanders

HC User, Elsa De Luca, Julia Flanders
2020
Conference proceedings of the Music Encoding Conference 2020 with Foreword by Richard Freedman and Anna J. Kijas.  ...  The new language, harmalysis, is based principally on Huron's **harm syntax, which was originally intended for accompanying music scores encoded in the Humdrum(**kern) representation.  ...  Recently, the interest for harmonic analysis and its standardization in machine-readable contexts has been revisited by academics [4] as well as developers of music notation software [6].  ... 
doi:10.17613/mvxw-x477 fatcat:475cce3hebagrn6j4ch2d4myqu

Discrimination of short speech‐like formant transitions

Astrid van Wieringen, Louis C. W. Pols
1992 Journal of the Acoustical Society of America  
The MAA data using headphones will be compared with free-field MAA data from the literature and with mean localization error data using headphones. 10:4S laPP9.  ...  The paradigm used was the two-source two-interval experiment described by Hartmann and Rakerd [J. Acoust. Soc. Am. aS, 2031-2041 ( 1985)].  ...  The position of the peaks varied with the paradigm.  ... 
doi:10.1121/1.405128 fatcat:dbtewzqvijhkta6op7ul3ribti

Dagstuhl Reports, Volume 9, Issue 1, January 2019, Complete Issue [article]

2019
The key feature of SLIONS is an automatic speech recognition (ASR) tool used for objective assessment of sung lyrics, which provides students with personalized, granular feedback based on their singing  ...  As an alternative, we suggest learning common mid-level representations for both query and database with a data-driven approach, e. g., using deep learning with the triplet loss [5] .  ...  Furthermore, I discussed how to automatically create large datasets annotated with pitch and lyrics using a student-teacher paradigm [4] .  ... 
doi:10.4230/dagrep.9.1 fatcat:m3grhk5hanccbg7oxkhos7kv4e
« Previous Showing results 1 — 15 out of 25 results