Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Look into datasets #11

Open
vvolhejn opened this issue Feb 28, 2022 · 4 comments
Open

Look into datasets #11

vvolhejn opened this issue Feb 28, 2022 · 4 comments

Comments

@vvolhejn
Copy link
Owner

No description provided.

@vvolhejn vvolhejn added this to the Reproduce baselines milestone Feb 28, 2022
@vvolhejn
Copy link
Owner Author

vvolhejn commented Mar 10, 2022

DDSP:

  • violin: uses five performances from Musopen, namely John Garner's numbers II, III, IV, VI, VII
  • NSynth: 300k four-second samples of MIDI notes, but the notes are separated which is unrealistic

Tone Transfer (paper):

  • trumpet and saxophone: "home recordings"
  • flute: performances from Musopen. "Four pieces by Paolo Damoro (24 Etudes for Flute, Op. 15 - III. Allegro conbrio in G major, 24 Etudes for Flute, Op. 15 - VI. Moderato in B minor, 3 Fantaisies for Solo Flute, Op. 38 - Fantaisie no. 1, Sonata Appassionata, Op. 140))"

NEWT:

RAVE:

  • strings: "an internal dataset composed of approximately 30 hours of raw recordings of strings in various configurations (monophonic solos and polyphonic group performances, with different styles and recording configurations)"
  • VCTK: "The Voice Conversion ToolKit (Yamagishi et al., 2019) is a speech dataset composed of approximately 44 hours of raw audio sampled at 48kHz, produced by 110 different speakers with various accents"

@vvolhejn
Copy link
Owner Author

https://www.ismir.net/resources/datasets/ ISMIR's list of datasets

@vvolhejn vvolhejn reopened this Mar 30, 2022
@vvolhejn
Copy link
Owner Author

From the ISMIR list, except the ones already mentioned above:

Good ones

VocalSet https://zenodo.org/record/1442513#.W7OaFBNKjx4

  • trained singers singing notes, scales and short melodies in different styles

GoodSounds https://www.upf.edu/web/mtg/good-sounds

  • recordings of notes and scales on classical instruments

GuitarSet https://guitarset.weebly.com/

  • guitars comping + soloing in different styles and keys

MARG-AMT http://marg.snu.ac.kr/automatic-music-transcription/

  • amateurs singing

Potentially useful

https://sigsep.github.io/datasets/musdb.html#musdb18-compressed-stems

  • not a single acoustic environment

https://gitlab.doc.gold.ac.uk/expressive-musical-gestures/dataset/-/tree/master/violin

  • one short melody snipped played by many players in different styles - maybe not diverse enough

http://mac.citi.sinica.edu.tw/GuitarTranscription/

  • single guitar notes with different techniques and amp settings

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant