collection of pitch detection algorithms with unified interface
Project description
pitch-detectors
collection of pitch detection algorithms with unified interface
list of algorithms
algorithm | cpu | gpu | accuracy [1] |
---|---|---|---|
PraatAC | ✓ | 0.880 | |
PraatCC | ✓ | 0.893 | |
PraatSHS | ✓ | 0.618 | |
Pyin | ✓ | 0.886 | |
Reaper | ✓ | 0.826 | |
Yaapt | ✓ | 0.759 | |
World | ✓ | 0.873 | |
TorchYin | ✓ | 0.886 | |
Rapt | ✓ | 0.859 | |
Swipe | ✓ | 0.871 | |
Crepe | ✓ | ✓ | 0.802 |
TorchCrepe | ✓ | ✓ | 0.817 |
Spice | ✓ | ✓ | 0.908 |
- [1] accuracy is mean raw pitch accuracy on 1000 samples of MIR-1K dataset
install
all agorithms tested on python3.10, this is recommended python version to use
pip install pitch-detectors
usage
from scipy.io import wavfile
from pitch_detectors import algorithms
import matplotlib.pyplot as plt
fs, a = wavfile.read('data/b1a5da49d564a7341e7e1327aa3f229a.wav')
pitch = algorithms.Crepe(a, fs)
plt.plot(pitch.t, pitch.f0)
plt.show()
additional features
- robust (vote-based + median) averaging of pitch
- json import/export
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
pitch-detectors-0.1.0.tar.gz
(7.3 kB
view hashes)
Built Distribution
Close
Hashes for pitch_detectors-0.1.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | df690a136ea4393667b12cad8d667284ddeff4f3c5d8c1e2a10a4b8b7942e6be |
|
MD5 | 393c0cded8eedcf262082fdcf432da98 |
|
BLAKE2b-256 | cf7257b00f662469ec2190ae0f5a5fa621b9bbf484ae91beda361b12d93fd75b |