GFDatabase: A Database of Flute Playing Techniques

GFDatabase (Geidai Flute Database) is a collection of recordings for the training of models for the real-time recognition of flute playing techniques.

The database includes recordings of two flute players, a professional and a semi-professional. It includes monophonic recordings of isolated pitches of 11 flute playing techniques: aeolian, flatterzunge, key click, multiphonics, ordinario, pizzicato, play and sing, staccato, tongue ram, trill, and whistle tone.

The professional player was recorded using seven microphones representing different typical studio performance recording sets (see table below) in studio A (a large recording studio) at Geidai, Tokyo University of the Arts. The recording was made using a 96kHz sampling rate and a resolution of 24 bits.

The semi-professional player was recorded using two microphones in studio C (a small recording studio) at Geidai, Tokyo University of the Arts. The recording was made using a 48kHz sampling rate and a resolution of 24 bits.

If you use our recordings in your work, please quote our paper:

Nicolas Brochec, Tsubasa Tanaka, Will Howie. Microphone-based Data Augmentation for Automatic Recognition of Instrumental Playing Techniques. International Computer Music Conference (ICMC 2024), Jul 2024, Seoul, South Korea.

This research is supported by the European Research Council (ERC) as part of the Raising Co-creativity in Cyber-Human Musicianship (REACH) Project directed by Gérard Assayag, under the European Union's Horizon 2020 research and innovation program (GA #883313). Funding support for this work was provided by a Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT) scholarship to Nicolas Brochec.