librosa

name	librosa
description	音频信号处理 - 重采样、STFT、梅尔频谱、恒Q变换特征提取
category	audio
emoji	🎵
github_url	https://github.com/librosa/librosa
triggers	["音频特征","频谱分析","梅尔频谱","librosa","特征提取"]
inputs	["wav","mp3","flac"]
outputs	["numpy","json"]
use_cases	["音频特征提取","频谱分析"]

Librosa

音频信号处理基础库。

安装

pip install librosa

使用

import librosa
y, sr = librosa.load('audio.wav')
mfccs = librosa.feature.mfcc(y=y, sr=sr)