SpecVQGAN

Visual sound generator

A project to train an audio generation model that uses visual cues to produce high-quality sounds from a reduced dataset of representative vectors.

Source code for "Taming Visually Guided Sound Generation" (Oral at the BMVC 2021)

GitHub

353 stars
8 watching
39 forks
Language: Jupyter Notebook
last commit: about 2 years ago
audioaudio-generationbmvcevaluation-metricsganmelganmulti-modalpytorchtransformervasvggsoundvideovideo-featuresvideo-understandingvqvae

Related projects:

RepositoryDescriptionStars
virtualanalogy/paraphrasisA spectral modeling synthesizer for generating sound.32
peihaochen/regnetAn implementation of a neural network for generating sound from video sequences52
tyorikan/voice-recording-visualizerA Java-based Android app for visualizing audio recordings550
l0sg/seqgan-musicGenerates polyphonic music sequences using deep learning models and adversarial training28
salu133445/museganGenerates polyphonic music from scratch or based on user input using a neural network architecture1,863
chrisdonahue/waveganAn open-source machine learning algorithm for generating raw audio waveforms from raw data1,334
mrugalla/nel-19An open-source plugin for generating vibrato effects in audio signals77
vincentherrmann/pytorch-wavenetAn implementation of WaveNet for generating audio using PyTorch979
nicolas-van/sonant-xA JavaScript synthesizer library used to generate sound effects and music for video games and small applications.236
mfcc64/youtube-musical-spectrumAn audio visualizer extension that displays spectrograms of YouTube videos and microphone input on web pages.175
ghostnan/revidia-audio-visualizerAn audio visualizer software that generates real-time visuals based on incoming audio streams24
dasetwas/enginesoundA software tool that generates synthetic engine sounds with customizable parameters and real-time preview capabilities.313
staskobzar/vue-audio-visualAn audio visualization plugin for the VueJS framework using HTML5 Web Audio API.726
viktornova/nauseaA music visualizer that uses audio spectrum analysis to create real-time visualizations of an audio stream.10
jonathanzwhite/audio-visualizerA program for visualizing audio signals in real-time.19