search

shaws34 / Audio-Undistortion

0 stars 0 forks source link

Find open source audio files #1

Open shaws34 opened 6 months ago

shaws34 commented 6 months ago

For training the model, I'll need human speech audio files. They should be clear and easy-to-understand files. Length and format will hopefully not be an issue. If there are public domain audio books, I think these could provide great training data.

shaws34 commented 6 months ago

https://www.digitalbook.io/audiobooks contains many free audio books which I think will make for good training data.

shaws34 commented 6 months ago

LibriSpeech provides a dataset - https://www.openslr.org/12

MarcoRavich commented 4 months ago

Hi there, we're a small italian nonprofit organization devoted to support indipendent/emerging artists mainly through audiovisuals content creations of their live performances.

We therefore have several hours of - both cameras' (= distorted and lossy) and main mixer's (= lossless and postproduced) - audio recordings from dozens shows we've documented over the years.

Let us know if these can be useful to evolve your project in live-music field.