当前位置:网站首页>Extraction de la fonction Speech de l'ensemble de données emo - DB
Extraction de la fonction Speech de l'ensemble de données emo - DB
2022-04-22 04:48:00 【Un vrai camarade de classe Cao.】
EMO-DB De l'ensemble de données Speech Extraction des caractéristiques
Introduction à l'ensemble de données
- EMO-DBL'ensemble de données est une base de données de voix émotionnelle en allemand enregistrée par l'Université technique de Berlin
- 535Fichiers vocaux(Il y a dix acteurs, dont cinq hommes et cinq femmes).Lorsque le contenu de l'énoncé contient des mots de la vie quotidienne5Une phrase courte et5Une longue phrase.,Avec une plus grande liberté émotionnelle,Ne contient pas une tendance émotionnelle particulière.Adoption16kHZÉchantillonnage,16bitQuantification,EtWAVFormat enregistrer le fichier.
- Dont la composition émotionnelle:anger, neutral, fear, boredom, happiness, sadness, disgust.

Principalement en utilisant cette étiquette vocale audio DocumentXXXWa.wavMoyenneWC'est l'émotion

Extraire les caractéristiques pertinentes
import feature
import os
import pickle
import tqdm
def extract_emodb(file_name:str):
label = file_name.split('.')[0][-2]
feature_27 = feature.extract_mfcc_feature(file_name)
return (label,feature_27)
database_dir = "H:\dataset\EM-DB\wav"
feature_tuple =()
dir_file = os.listdir(database_dir)
for i in tqdm.tqdm(dir_file):
file_path = os.path.join(database_dir,i)
feature_tuple = feature_tuple+ (extract_emodb(file_path),)
with open('emo_db.pkl', 'wb+') as file:
pickle.dump(feature_tuple,file)
Outils
def read_file(filename):
""" input a filename and get wave data and time,framerate :param filename: :return: """
file = wave.open(filename, 'r')
params = file.getparams()
nchannels, sampwidth, framerate, wav_length = params[:4]
str_data = file.readframes(wav_length)
wavedata = np.fromstring(str_data, dtype=np.short)
time = np.arange(0, wav_length) * (1.0 / framerate)
file.close()
return wavedata, time, framerate
def extract_mfcc_feature(filename):
""" input a filename extract mfcc from audio :param filename: :return: """
data, time, rate = read_file(filename)
mel_spec = ps.logfbank(data, rate, nfilt=13)
np.set_printoptions()
time = mel_spec.shape[0]
if time <= 300:
part = mel_spec
part = np.pad(part, ((0, 300 - part.shape[0]), (0, 0)), 'constant',
constant_values=0)
mfcc = dct(part)
else:
begin = 0
end = 300
part = mel_spec[begin:end, :]
mfcc = dct(part)
return mfcc
Résumé
Cet ensemble de données est encore relativement simple ,C'est clair.,Je ne comprends pas..
版权声明
本文为[Un vrai camarade de classe Cao.]所创,转载请带上原文链接,感谢
https://yzsam.com/2022/04/202204220447377169.html
边栏推荐
- Sharing of enterprise data leakage prevention scheme
- Iris species prediction -- Introduction to data set
- Carina 的根基與誕生背景|深入了解 Carina 系列 第一期
- 2021-08-14
- [selenium] basic application of unittest testing framework
- 10.libevent接收和处理服务器消息
- 11. Libevent tests horizontal trigger and edge trigger
- [chestnut sugar GIS] ArcMap - how to combine multiple images into one
- C-handwritten 10
- 使用指定显卡
猜你喜欢

Opencv usage record of neural network learning

Vue project NPM run build when packaging the project, time stamp the version number of CSS and JS files to prevent the browser from caching
![[SQL Server accelerated path] database and table (II)](/img/5a/cb54c5ca894820ab46b02d4c0ee346.png)
[SQL Server accelerated path] database and table (II)

【Selenium】UnitTest测试框架的基本应用

vue项目 npm run build 打包项目时为css、js文件加时间戳版本号,防止浏览器缓存

Carina local storage selected into the CNCF cloud native panorama

JVM - G1 garbage collector

Linear regression API

Feature Engineering - feature preprocessing (normalization, standardization)

goland汉化解决方法(下载插件失败的情况下)
随机推荐
Boyun beyondcmp cloud management platform version 5.6 release
Common commands of labelme
10.libevent接收和处理服务器消息
Opencv usage record of neural network learning
Discussion on data set division
Win10 Caton repair
crypto-js加密算法库【安装教程、缓存加密】
LeetCode 剑指 Offer 22. 链表中倒数第k个节点
Raspberry pie 4B compile paddlelite (detailed steps 2022)
Target detection - lightweight network (as of April 21, 2022)
WAP picture
Paper reading (48): a library of optimization algorithms for organizational design
First knowledge of UI automation (inspect.exe + uiautomation)
[selenium] basic application of unittest testing framework
Carina local storage selected into the CNCF cloud native panorama
仿真生成随机数计算生成每个同学生日
一个基于.NET Core3.1的开源项目帮你彻底搞懂WPF框架Prism
Deployment of web server, personal experience
How to realize different values and display different colors in MATLAB
Leetcode sword finger offer 17 Print from 1 to maximum n digits