audio2numpy load an audio file and directly ouputs the audio data as a numpy array and its sampling rate. Supports .wav, .aiff via python's standard library, and other formats via ffmpeg.
Using pip:
pip install git+https://github.com/irgendwr/audio2numpy
audio2numpy requires ffmpeg to decode mp3 files. You would need to install ffmpeg in order to have mp3 support.
homebrew install ffmpeg
- Debian/Ubuntu:
sudo apt-get install ffmpeg - Arch:
sudo pacman -S ffmpeg
Check here for other installation methods for different Linux distributions.
- Download ffmpeg
- Extract it into a folder, for example
C:\FFmpeg - Add the ffmpeg bin folder to your PATH Environment Variable.
Here is a guide that explains the process in detail.
!apt install ffmpeg
from audio2numpy import open_audio
path = "./examples/word.mp3"
signal, sampling_rate = open_audio(path)0.1.4 (09.09.2021)
- Remove format limitation: wiccy46/audio2numpy#2
0.1.3 (17.10.2019)
- Add Colab ffmpeg installation guide.
- Remove ffmpeg from requirements.txt. Users should install it separately
0.1.2 (20.08.2019)
Add instructions to install ffmpeg if load mp3 failed with ffmpeg backend not available.
0.1.1 (14.08.2019)
Initial release.