datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
OpenS2S_Datasets
How to Use?
Download, merge the files, and extract
You can run the following command to merge the compressed file parts after downloading.
cat en_response_wav.tar.gz.* > en_response_wav.tar.gz
cat zh_response_wav.tar.gz.* > zh_response_wav.tar.gz
lmd_mp3
The Lakh MIDI Dataset in MP3
The MIDI files from LMD were synthesized and split into 15-second segments.
Soundfont: GeneralUser GS 2.02
Notes:
Files that were skipped:
Corrupt or unreadable by pretty_midi or mido.
Could not be synthesized with fluidsynth.
Longer than 20 minutes.
(Some) ending segments shorter than 5 seconds.
Reference:
Colin Raffel. "Learning-Based Methods for Comparing Sequences, with Applications to Audio-to-MIDI Alignment and Matching". PhD Thesis, 2016.