Search Results for author: Keisuke Toyama

Found 3 papers, 2 papers with code

Timbre-Trap: A Low-Resource Framework for Instrument-Agnostic Music Transcription

no code implementations27 Sep 2023 Frank Cwitkowitz, Kin Wai Cheuk, Woosung Choi, Marco A. Martínez-Ramírez, Keisuke Toyama, Wei-Hsiang Liao, Yuki Mitsufuji

Several works have explored multi-instrument transcription as a means to bolster the performance of models on low-resource tasks, but these methods face the same data availability issues.

Music Transcription

Automatic Piano Transcription with Hierarchical Frequency-Time Transformer

1 code implementation10 Jul 2023 Keisuke Toyama, Taketo Akama, Yukara Ikemiya, Yuhta Takida, Wei-Hsiang Liao, Yuki Mitsufuji

This is especially helpful when determining the precise onset and offset for each note in the polyphonic piano content.

Music Transcription

E2E Refined Dataset

1 code implementation1 Nov 2022 Keisuke Toyama, Katsuhito Sudoh, Satoshi Nakamura

Although the well-known MR-to-text E2E dataset has been used by many researchers, its MR-text pairs include many deletion/insertion/substitution errors.

Cannot find the paper you are looking for? You can Submit a new open access paper.