Beethoven Piano Sonata Audio Waveforms for Music Generation

Raw audio waveforms of single-instrument piano music, specifically Beethoven sonatas. It was introduced in the SampleRNN paper by Mehri et al. (2017) and later used for training music generation models in the paper 'It's Raw! Audio Generation with State-Space Models'.

Use Cases

• Train state-space models on raw audio waveforms for small-scale music generation.
• Benchmark audio generation models using single-instrument piano music from Beethoven sonatas.
• Analyze the structure of raw audio waveforms from classical piano compositions.

Strengths

• Dataset is focused on single-instrument piano music, providing a clean audio source.
• Serves as a benchmark, having been used in published research papers on audio generation.

Limitations

• The dataset size is described as 'small scale', suggesting a limited number of audio samples.
• Specific details on the number of audio files, duration, or sampling rate are not provided in the input.

Provenance

Source

Originally introduced in the SampleRNN paper by Mehri et al. (2017).

Collection Method

null

Time Range

null

Freshness

Last updated on 2022-02-22.

Geography

null