Releases: BirdVox/birdvoxpaint
0.2.2
This version is the second bugfix of the 0.2.X series.
Following librosa PR #1431 (merged in 2022), we now call librosa functions with explicit keyword arguments. Hence, we support librosa v0.9+
The behavior of birdvoxpaint is unchanged and remains backwards compatible with earlier versions of librosa.
0.2.1
0.2.0
This is the second stable release of BirdVoxPaint.
The main new functionalities are:
- support for mel-spectrograms
- progress bar with tqdm
- RGB display:
rgbshow - new acoustic indices, based on Per-Channel Energy Normalization (PCEN)
BirdVoxPaint is financially supported by the BirdVox NSF grant and the Pinkerton Foundation. Its core developers are Vincent Lostanlen (Cornell Lab of Ornithology), Phincho Sherpa (New York University), and Bea Steers (New York University).
v0.1.0
BirdVoxPaint is a Python module for computing spectrotemporal indices of acoustic complexity in long-duration recordings.
The initial release (v0.1.0) offers five of these indices:
- acoustic complexity index (ACI) of Farina and Pieretti
- average energy
- entropy-based concentration of Sueur et al.
- maximum energy
- maximum spectral flux (without logarithmic compression before differentiation)
The birdvoxpaint.transform function computes these indices efficiently while guaranteeing an upper bound on the memory footprint. By default, it parallelizes across all available cores.
BirdVoxPaint is financially supported by the BirdVox NSF grant and the Pinkerton Foundation. Its core developers are Vincent Lostanlen (Cornell Lab of Ornithology) and Phincho Sherpa (New York University). The only dependency of the birdvoxpaint module is librosa>=0.7.0.
In the present version, birdvoxpaint only support STFT-based indices. Support for melspectrogram-based indices is a planned feature.