Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 9 of 9 for “"Music Source Separation"”.

  1. Music Visualization Using Source Separated Stereophonic Music

    This thesis introduces a music visualization system for stereophonic source separated music. Music visualization systems are a popular way to represent information from audio signals through computer graphics. Visualization can help people better understand music and its complex and interacting …

    calpoly Repository record for Music Visualization Using Source Separated Stereophonic Music (opens in a new tab)

  2. Deep learning approaches for automatic sung speech recognition: adapting spoken technologies to sung speech

    … the difficulty of separating the vocals from the musical accompaniment. In recent years, deep neural network techniques have revolutionised spoken speech recognition systems through advances in both acoustic modelling and audio source separation. This thesis evaluates whether these new techniques …

    whiterose Download PDF for Deep learning approaches for automatic sung speech recognition: adapting spoken technologies to sung speech (opens in a new tab)

  3. Unsupervised Detection and Synthesis of Speech and Environmental Sounds Using Generative Networks

    … individual instruments or vocals in a piece of music. This isn’t unique to just humans, as most animals capable of hearing need to focus on specific environmental noises for their own safety, whether that’s localizing where a branch just broke warning of an encroaching predator or tracking the …

    whiterose Download PDF for Unsupervised Detection and Synthesis of Speech and Environmental Sounds Using Generative Networks (opens in a new tab)

  4. Musical instrument sound source separation

    The structured arrangement of sounds in musical pieces, results in the unique creation of complex acoustic mixtures. The analysis of these mixtures, with the objective of estimating the individual sounds which constitute them, is known as musical instrument sound source separation, and has …

    unsw Repository record for Musical instrument sound source separation (opens in a new tab)

  5. Monaural Music Separation via Supervised Non-Negative Matrix Factor with Side-Information

    In this dissertation, a supervised source template nonnegative matrix factorization (NMF) algorithm is proposed to solve the monaural music source separation problem. Different from the previous state-of-the-art algorithms, the basic theoretical concept of the proposed algorithm considers the …

    carleton Repository record for Monaural Music Separation via Supervised Non-Negative Matrix Factor with Side-Information (opens in a new tab)

  6. Musical source separation with deep learning and large-scale datasets

    Throughout this thesis we will explore automatic music source separation by utilizing modern (at the time of writing) techniques and tools from machine learning and big data processing. The bulk of this work was carried out between 2016 and 2019. In Chapter 2 we conduct a review of source …

    city-london Download PDF for Musical source separation with deep learning and large-scale datasets (opens in a new tab)

  7. Aplikace pro separaci kytarového zvuku z hudební nahrávky

    Cieľom tejto práce bolo implementovať model na separáciu gitarového zvuku z nahrávky a použiť ho v praktickej aplikácii. Bolo nutné manuálne vytvoriť vlastný trénovací dataset z remixov piesní a upraviť existujúci MedleyDB dataset pre naše účely. Ako základ neurónovej siete sme si vybrali Demucs …

    brno-tech Repository record for Aplikace pro separaci kytarového zvuku z hudební nahrávky (opens in a new tab)

  8. Separace nástrojů a zpěvu z hudební nahrávky

    Na separáciu zdrojov z hudobných nahrávok sa používajú neurónové siete. Jednou z takýchto sietí je Conv-TasNet. Cieľom práce je experimentovať s už existujúcou implementáciou tejto siete za účelom potenciálneho zlepšenia. Trénovanie modelov prebiehalo na dátovej sade MUSDB18. Postupne sa …

    brno-tech Repository record for Separace nástrojů a zpěvu z hudební nahrávky (opens in a new tab)

  9. Strojové učení v audio efektech

    Získávání hudebních efektů z mixovaných skladeb je složité téma, které vyžaduje znalosti jak v oblasti zpracování signálů, tak zkušenosti s audio inženýrstvím. Tato práce cílí na tvorbu systém, který by byl schopen identifikovat sekvence a parametry kytarových efektů z mixovaných skladeb. Trénovací …

    brno-tech Repository record for Strojové učení v audio efektech (opens in a new tab)