Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/csteinmetz1/st-ito

Audio production style transfer with inference-time optimization

https://github.com/csteinmetz1/st-ito

Science Score: 26.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (5.0%) to scientific vocabulary
Last synced: 9 months ago · JSON representation

Repository

Audio production style transfer with inference-time optimization

Basic Info
Statistics
  • Stars: 41
  • Watchers: 1
  • Forks: 4
  • Open Issues: 1
  • Releases: 0
Created about 2 years ago · Last pushed over 1 year ago
Metadata Files
Readme License

README.md

st-ito

Audio production style transfer with inference-time optimization

Quick Start

Setup Bash git clone cd st-ito pip install -e .

Running style transfer on an audio file Bash python scripts/run_optim.py \ "input.wav" \ --target "target.wav" \ --algorithm es \ --effect-type vst \ --dropout 0.0 \ --max-iters 25 \ --metric param

Using AFx-Rep to extract embeddings ```Python import torch import torchaudio from stito.utils import loadparammodel, getparam_embeds

load pretrained model

model = loadparammodel(use_gpu=True)

load audio file

audio, sr = torchaudio.load("input.wav")

audio must be of shape bs, chs, seq_len

audio = audio.unsqueeze(0)

extract embeddings

embeddict = getparamembeds(audio, model, sr) for embedname, embed in embeddict.items(): print(embedname, embed.shape)

mid torch.Size([1, 512])

side torch.Size([1, 512])

```

Training

```

```

Owner

  • Name: Christian J. Steinmetz
  • Login: csteinmetz1
  • Kind: user
  • Location: London, UK
  • Company: @aim-qmul

Machine learning for Hi-Fi audio. PhD Researcher at C4DM.

GitHub Events

Total
  • Issues event: 2
  • Watch event: 39
  • Issue comment event: 4
  • Push event: 2
  • Public event: 1
  • Pull request event: 1
  • Fork event: 4
Last Year
  • Issues event: 2
  • Watch event: 39
  • Issue comment event: 4
  • Push event: 2
  • Public event: 1
  • Pull request event: 1
  • Fork event: 4

Issues and Pull Requests

Last synced: over 1 year ago

All Time
  • Total issues: 1
  • Total pull requests: 1
  • Average time to close issues: 4 days
  • Average time to close pull requests: N/A
  • Total issue authors: 1
  • Total pull request authors: 1
  • Average comments per issue: 4.0
  • Average comments per pull request: 0.0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 1
  • Pull requests: 1
  • Average time to close issues: 4 days
  • Average time to close pull requests: N/A
  • Issue authors: 1
  • Pull request authors: 1
  • Average comments per issue: 4.0
  • Average comments per pull request: 0.0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
  • sonovice (1)
Pull Request Authors
  • ntamotsu (1)
Top Labels
Issue Labels
Pull Request Labels

Dependencies

setup.py pypi
  • auraloss *
  • cma *
  • dasp-pytorch *
  • laion_clap *
  • matplotlib *
  • pedalboard *
  • pyloudnorm *
  • pytorch-lightning *
  • resampy *
  • scikit-learn *
  • soundfile *
  • timm *
  • torch *
  • torchaudio *
  • torchlibrosa *
  • torchvision *
  • transformers *
  • umap-learn *
  • wandb *
  • wav2clip *