A Semi-Automatic Application to Segment and Tag Recordings from Vinyls.

C. Junker, Peter Knees

Research output: Chapter in Book/Report/Conference proceedingConference proceedingspeer-review

Abstract

Digitisation of analogue audio storage media, such as vinyls, can be tedious work. Transferring the contained music to, e.g., an mp3 file that can be put on portable music devices usually requires several steps that exceed the (already not trivial) process of recording the pure audio signal. As opposed to ripping a CD, where audio extraction and labelling with meta-data can be performed fully automatic, for vinyls, the user has to take care of flipping the disc to record both sides, finding start and end of every track, extracting individual tracks, and finally manually annotating each track with meta-data. Disc-JoQey is an open-source software (published under BSD license) that aims at supporting the user in the process of segmenting, extracting, and tagging tracks from vinyl recordings. After typing in album artist and album title, Disc-JoQey retrieves album-specific meta-data (track titles, lengths, and offsets) via a music meta-data search engine (currently MusicBrainz). Since the retrieved data refers to the CD version of the album, timing data has to be adapted (i.e., mapped to the two sides of the recording). A waveform representation of the recorded signals can be used to check the correctness of start and end point for each track and, if necessary, to correct them. All annotated tracks can then be automatically exported into id3-tagged mp3 files.
Original languageEnglish
Title of host publicationDemonstration Session. 10th International Conference on Music Information Retrieval (ISMIR 2009)
Number of pages1
Publication statusPublished - 2009

Fields of science

  • 102 Computer Sciences
  • 102001 Artificial intelligence
  • 102003 Image processing
  • 102015 Information systems
  • 202002 Audiovisual media

Cite this